Event-Based Vision: A Survey

"Seeing" ENF From Neuromorphic Events: Modeling and Robust Estimation

Most artificial lights exhibit subtle fluctuations in intensity and frequency in response to the influence of the grid's alternating current, providing the potential to estimate the Electric Network Frequency (ENF) from conventional frame-based videos. Nevertheless, the performance of Video-based ENF (V-ENF) estimation largely relies on the imaging quality and thus may suffer from significant interference caused by non-ideal sampling, scene diversity, motion interference, and extreme lighting conditions. In this paper, we show that the ENF can be extracted without the above limitations from a new modality provided by the so-called event camera, a neuromorphic sensor that encodes the light intensity variations and asynchronously emits events with extremely high temporal resolution and high dynamic range. Specifically, we formulate and validate the physical mechanism for the ENF captured in events and then propose a simple yet robust Event-based ENF (E-ENF) estimation method through mode filtering and harmonic enhancement. To validate the effectiveness, we build the first Event-Video ENF Dataset (EV-ENFD) and its extension EV-ENFD+ with diverse scenarios, including static, dynamic, and extreme lighting scenes. Comprehensive experiments have been conducted on our proposed datasets, showcasing that our proposed E-ENF significantly outperforms the V-ENF in extracting accurate ENF traces, especially in challenging environments.

A Biomimetic Visual Detection Model: Event-Driven LGMDs Implemented With Fractional Spiking Neuron Circuits

OBJECTIVE

Lobula giant motion detectors (LGMDs) in locusts effectively predict collisions and trigger avoidance, with potential applications in autonomous driving and UAVs. Research on LGMD characteristics splits into two views: one focusing on the presynaptic visual pathway, the other on the postsynaptic LGMD neurons. Both perspectives have support, leading to two computational models, but they lack a biophysical description of the individual LGMD neuron behavior. This paper aims to mimic and explain LGMD behavior based on fractional spiking neurons (FSNs) and construct a biomimetic visual model for the LGMD compatible with these characteristics.

METHODS

We implement the visual model using an event camera to simulate photoreceptors and follow the ON/OFF visual pathway, incorporating lateral inhibition to mimic the LGMD system from the bottom up. Second, most computational models of motion perception use only the dendrites within the LGMD neurons as the ideal pathway for linear summation, ignoring dendritic effects inducing neuronal properties. Thus, we introduce FSN circuits by altering dendritic morphological parameters to simulate multi-scale spike frequency adaptation (SFA) observed in LGMDs. Additionally, we add one more circuit of dendritic trees into the FSNs to be compatible with the postsynaptic feed-forward inhibition (FFI) in LGMD neurons, providing a novel explanatory and predictive model.

RESULTS

We test that the event-driven biomimetic visual model can achieve collision detection and looming selection in different complex scenes, especially fast-moving objects.

SpQuant-SNN: ultra-low precision membrane potential with sparse activations unlock the potential of on-device spiking neural networks applications

Spiking neural networks (SNNs) have received increasing attention due to their high biological plausibility and energy efficiency. The binary spike-based information propagation enables efficient sparse computation in event-based and static computer vision applications. However, the weight precision and especially the membrane potential precision remain as high-precision values (e.g., 32 bits) in state-of-the-art SNN algorithms. Each neuron in an SNN stores the membrane potential over time and typically updates its value in every time step. Such frequent read/write operations of high-precision membrane potential incur storage and memory access overhead in SNNs, which undermines the SNNs' compatibility with resource-constrained hardware. To resolve this inefficiency, prior works have explored the time step reduction and low-precision representation of membrane potential at a limited scale and reported significant accuracy drops. Furthermore, while recent advances in on-device AI present pruning and quantization optimization with different architectures and datasets, simultaneous pruning with quantization is highly under-explored in SNNs. In this work, we present SpQuant-SNN, a fully-quantized spiking neural network with ultra-low precision weights, membrane potential, and high spatial-channel sparsity, enabling the end-to-end low precision with significantly reduced operations on SNN. First, we propose an integer-only quantization scheme for the membrane potential with a stacked surrogate gradient function, a simple-yet-effective method that enables the smooth learning process of quantized SNN training. Second, we implement spatial-channel pruning with membrane potential prior, toward reducing the layer-wise computational complexity, and floating-point operations (FLOPs) in SNNs. Finally, to further improve the accuracy of low-precision and sparse SNN, we propose a self-adaptive learnable potential threshold for SNN training. Equipped with high biological adaptiveness, minimal computations, and memory utilization, SpQuant-SNN achieves state-of-the-art performance across multiple SNN models for both event-based and static image datasets, including both image classification and object detection tasks. The proposed SpQuant-SNN achieved up to 13× memory reduction and >4.7× FLOPs reduction with < 1.8% accuracy degradation for both classification and object detection tasks, compared to the SOTA baseline.

Artificial Amacrine Retinal Circuits

Event-based imaging represents a new paradigm in visual information processing that addresses the speed and energy efficiency shortcomings inherently present in the current complementary metal oxide semiconductor-based machine vision. Realizing such imaging systems has previously been sought using very large-scale integration technologies that have complex circuitries consisting of many photodiodes, differential amplifiers, capacitors, and resistors. Here, we demonstrate that event-driven sensing can be achieved using a simple one-resistor, one-capacitor (1R1C) circuit, where the capacitor is modified with colloidal quantum dots (CQDs) to have a photoresponse. This sensory circuit emulates the motion-tracking function of the biological retina, in which the amacrine cells in the bipolar-to-ganglion synaptic pathway produce a transient spiking signal only in response to changes in light intensity but remain inactive under constant illumination. When extended to a 2D imaging array, the individual sensors work independently and output signals only when a change in the light intensity is detected; hence, the concept of the frame in image processing is thereby removed. In this work, we present the fabrication and characterization of a CQD photocapacitor-based 1R1C circuit that has a spectral response at 1550 nm in the short-wave infrared (SWIR). We report on the key performance parameters including peak responsivity, noise, and optical noise equivalent power and discuss the operating mechanism that is responsible for spiking responses in these artificial retinal circuits. The present work sets the foundation for expanding the bioinspired vision sensor capability toward midwave infrared (MWIR) and long-wave infrared (LWIR) spectral regions that are invisible to human eyes and mainstream semiconductor technologies.

Probing diffusive media through speckle differencing

Temporally varying speckle patterns, produced by light-matter interaction encode valuable information about inhomogeneities embedded within a scattering medium. These speckle fluctuations arise either from the tuning of the emission frequency of a laser illuminating a static scattering medium or from the microscopic motion of scatterers within a dynamically scattering medium. In this work, we detect embedded inhomogeneities by probing static and dynamic scattering media with coherent light and leveraging the statistical distribution of temporal speckle differences. In addition, we utilize the insights from the speckle differencing paradigm, to present the first experimental results of detecting inhomogeneities embedded within a scattering medium using bio-inspired neuromorphic sensors. The proposed neuromorphic approach simplifies the optical and electronic design, and significantly reduces data throughput by capturing only the differential information in the form of 1-bit spikes.

Negative Photoconductivity Transistors for Visuomorphic Computing

Visuomorphic computing aims to simulate and potentially surpass the human retina by mimicking biological visual perception with an artificial retina. Despite significant progress, challenges persist in perceiving complex interactive environments. Negative photoconductivity transistors (NPTs) mimic synaptic behavior by achieving adjustable positive photoconductivity (PPC) and negative photoconductivity (NPC), simulating "excitation" and "inhibition" akin to sensory cell signals. In complex interactive environments, NPTs are desired for visuomorphic computing that can achieve a better sense of information, lower power consumption, and reduce hardware complexity. In this review, it is started by introducing the development process of NPTs, while placing a strong emphasis on the device structures, working mechanisms, and key performance parameters. The common material systems employed in NPTs based on their functions are then summarized. Moreover, it is proceeded to summarize the noteworthy applications of NPTs in optoelectronic devices, including advanced multibit nonvolatile memory, optoelectronic logic gates, optical encryption, and visual perception. Finally, the challenges and prospects that lie ahead in the ongoing development of NPTs are addressed, offering valuable insights into their applications in optoelectronics and a comprehensive understanding of their significance.

Line-Based 6-DoF Object Pose Estimation and Tracking With an Event Camera

Pose estimation and tracking of objects is a fundamental application in 3D vision. Event cameras possess remarkable attributes such as high dynamic range, low latency, and resilience against motion blur, which enables them to address challenging high dynamic range scenes or high-speed motion. These features make event cameras an ideal complement over standard cameras for object pose estimation. In this work, we propose a line-based robust pose estimation and tracking method for planar or non-planar objects using an event camera. Firstly, we extract object lines directly from events, then provide an initial pose using a globally-optimal Branch-and-Bound approach, where 2D-3D line correspondences are not known in advance. Subsequently, we utilize event-line matching to establish correspondences between 2D events and 3D models. Furthermore, object poses are refined and continuously tracked by minimizing event-line distances. Events are assigned different weights based on these distances, employing robust estimation algorithms. To evaluate the precision of the proposed methods in object pose estimation and tracking, we have devised and established an event-based moving object dataset. Compared against state-of-the-art methods, the robustness and accuracy of our methods have been validated both on synthetic experiments and the proposed dataset. The source code is available at https://github.com/Zibin6/LOPET.

A Single-Pixel Event Photoactive Device for Real-Time, In-Sensor Spatiotemporal Optical Information Processing

The increasing demand for energy-efficient, sophisticated optical sensing technologies in various applications, from machine vision to optical communication, highlights the necessity for innovations in spatiotemporal information sensing and processing at a nearly single-pixel scale. Traditional methods, including multi-pixel photodetector arrays and event-based camera systems, often fail to provide rapid, real-time detection and processing of dynamic events within the sensor. This shortcoming is particularly notable in handling high-dimensional spatiotemporal data, where the dependency on sequential data input and external processing tools leads to latency, reduced throughput, and heightened energy consumption, thereby impeding real-time parallel data processing capabilities. Here, a carrier-selective, single-pixel, position-sensitive planar photoactive device that integrates spatiotemporal event sensing with inherent short-term memory capabilities is introduced. The proof-of-concept single-pixel event photoactive device enables in-sensor spatiotemporal parallel optical information processing, efficiently managing multibit (>4 bit) data simultaneously and facilitating ultrafast (≈0.4 µs) recognition of input patterns with low energy consumption (25 femtojoules). Additionally, by adjusting the operating speed from continuous to pulsed light illumination, the sensor array can detect trajectories and absolute position of events, offering in-sensor optical flow detection. This single-pixel event photodetector marks significant advancement toward developing compact, energy-efficient, ultrafast sensors suitable for a wide range of in sensor-based photonic applications.

High-performance deep spiking neural networks with 0.3 spikes per neuron

Communication by rare, binary spikes is a key factor for the energy efficiency of biological brains. However, it is harder to train biologically-inspired spiking neural networks than artificial neural networks. This is puzzling given that theoretical results provide exact mapping algorithms from artificial to spiking neural networks with time-to-first-spike coding. In this paper we analyze in theory and simulation the learning dynamics of time-to-first-spike-networks and identify a specific instance of the vanishing-or-exploding gradient problem. While two choices of spiking neural network mappings solve this problem at initialization, only the one with a constant slope of the neuron membrane potential at threshold guarantees the equivalence of the training trajectory between spiking and artificial neural networks with rectified linear units. For specific image classification architectures comprising feed-forward dense or convolutional layers, we demonstrate that deep spiking neural network models can be effectively trained from scratch on MNIST and Fashion-MNIST datasets, or fine-tuned on large-scale datasets, such as CIFAR10, CIFAR100 and PLACES365, to achieve the exact same performance as that of artificial neural networks, surpassing previous spiking neural networks. Our approach accomplishes high-performance classification with less than 0.3 spikes per neuron, lending itself for an energy-efficient implementation. We also show that fine-tuning spiking neural networks with our robust gradient descent algorithm enables their optimization for hardware implementations with low latency and resilience to noise and quantization.

Tailoring Space-Time Nonlocality for Event-Based Image Processing Metasurfaces

Analog computation with passive optical components can enhance processing speeds and reduce power consumption, recently attracting renewed interest thanks to the opportunities enabled by metasurfaces. Basic image processing tasks, such as spatial differentiation, have been recently demonstrated based on engineered nonlocalities in metasurfaces, but next-generation computational schemes require more advanced capabilities. Here, by simultaneously tailoring the nonlocal electromagnetic response of a metasurface in space and time, we demonstrate a passive ultrathin silicon-based device that performs mixed spatiotemporal differentiation of input images, realizing event-based edge detection. The metasurface performs spatial differentiation only when the input image is evolving in time, resulting in spatiotemporal image processing on subpicosecond timescales. Moreover, the metasurface design can be tailored to selectively enhance objects moving at desired speeds. Our results point towards fully passive processing of spatiotemporal signals, for highly compact neuromorphic cameras.

SiamEFT: adaptive-time feature extraction hybrid network for RGBE multi-domain object tracking

Integrating RGB and Event (RGBE) multi-domain information obtained by high-dynamic-range and temporal-resolution event cameras has been considered an effective scheme for robust object tracking. However, existing RGBE tracking methods have overlooked the unique spatio-temporal features over different domains, leading to object tracking failure and inefficiency, especally for objects against complex backgrounds. To address this problem, we propose a novel tracker based on adaptive-time feature extraction hybrid networks, namely Siamese Event Frame Tracker (SiamEFT), which focuses on the effective representation and utilization of the diverse spatio-temporal features of RGBE. We first design an adaptive-time attention module to aggregate event data into frames based on adaptive-time weights to enhance information representation. Subsequently, the SiamEF module and cross-network fusion module combining artificial neural networks and spiking neural networks hybrid network are designed to effectively extract and fuse the spatio-temporal features of RGBE. Extensive experiments on two RGBE datasets (VisEvent and COESOT) show that the SiamEFT achieves a success rate of 0.456 and 0.574, outperforming the state-of-the-art competing methods and exhibiting a 2.3-fold enhancement in efficiency. These results validate the superior accuracy and efficiency of SiamEFT in diverse and challenging scenes.

A general model unifying the adaptive, transient and sustained properties of ON and OFF auditory neural responses

Sounds are temporal stimuli decomposed into numerous elementary components by the auditory nervous system. For instance, a temporal to spectro-temporal transformation modelling the frequency decomposition performed by the cochlea is a widely adopted first processing step in today's computational models of auditory neural responses. Similarly, increments and decrements in sound intensity (i.e., of the raw waveform itself or of its spectral bands) constitute critical features of the neural code, with high behavioural significance. However, despite the growing attention of the scientific community on auditory OFF responses, their relationship with transient ON, sustained responses and adaptation remains unclear. In this context, we propose a new general model, based on a pair of linear filters, named AdapTrans, that captures both sustained and transient ON and OFF responses into a unifying and easy to expand framework. We demonstrate that filtering audio cochleagrams with AdapTrans permits to accurately render known properties of neural responses measured in different mammal species such as the dependence of OFF responses on the stimulus fall time and on the preceding sound duration. Furthermore, by integrating our framework into gold standard and state-of-the-art machine learning models that predict neural responses from audio stimuli, following a supervised training on a large compilation of electrophysiology datasets (ready-to-deploy PyTorch models and pre-processed datasets shared publicly), we show that AdapTrans systematically improves the prediction accuracy of estimated responses within different cortical areas of the rat and ferret auditory brain. Together, these results motivate the use of our framework for computational and systems neuroscientists willing to increase the plausibility and performances of their models of audition.

Millisecond-scale behaviours of plankton quantified in vitro and in situ using the Event-based Vision Sensor

The Event-based Vision Sensor (EVS) is a bio-inspired sensor that captures detailed motions of objects, aiming to become the 'eyes' of machines like self-driving cars. Compared to conventional frame-based image sensors, the EVS has an extremely fast motion capture equivalent to 10,000-fps even with standard optical settings, plus high dynamic ranges for brightness and also lower consumption of memory and energy. Here, we developed 22 characteristic features for analysing the motions of aquatic particles from the EVS raw data and tested the applicability of the EVS in analysing plankton behaviour. Laboratory cultures of six species of zooplankton and phytoplankton were observed, confirming species-specific motion periodicities up to 41 Hz. We applied machine learning to automatically classify particles into four categories of zooplankton and passive particles, achieving an accuracy up to 86%. At the in situ deployment of the EVS at the bottom of Lake Biwa, several particles exhibiting distinct cumulative trajectory with periodicities in their motion (up to 16 Hz) were identified, suggesting that they were living organisms with rhythmic behaviour. We also used the EVS in the deep sea, observing particles with active motion and periodicities over 40 Hz. Our application of the EVS, especially focusing on its millisecond-scale temporal resolution and wide dynamic range, provides a new avenue to investigate organismal behaviour characterised by rapid and periodical motions. The EVS will likely be applicable in the near future for the automated monitoring of plankton behaviour by edge computing on autonomous floats, as well as quantifying rapid cellular-level activities under microscopy.

Application of Event Cameras and Neuromorphic Computing to VSLAM: A Survey

Simultaneous Localization and Mapping (SLAM) is a crucial function for most autonomous systems, allowing them to both navigate through and create maps of unfamiliar surroundings. Traditional Visual SLAM, also commonly known as VSLAM, relies on frame-based cameras and structured processing pipelines, which face challenges in dynamic or low-light environments. However, recent advancements in event camera technology and neuromorphic processing offer promising opportunities to overcome these limitations. Event cameras inspired by biological vision systems capture the scenes asynchronously, consuming minimal power but with higher temporal resolution. Neuromorphic processors, which are designed to mimic the parallel processing capabilities of the human brain, offer efficient computation for real-time data processing of event-based data streams. This paper provides a comprehensive overview of recent research efforts in integrating event cameras and neuromorphic processors into VSLAM systems. It discusses the principles behind event cameras and neuromorphic processors, highlighting their advantages over traditional sensing and processing methods. Furthermore, an in-depth survey was conducted on state-of-the-art approaches in event-based SLAM, including feature extraction, motion estimation, and map reconstruction techniques. Additionally, the integration of event cameras with neuromorphic processors, focusing on their synergistic benefits in terms of energy efficiency, robustness, and real-time performance, was explored. The paper also discusses the challenges and open research questions in this emerging field, such as sensor calibration, data fusion, and algorithmic development. Finally, the potential applications and future directions for event-based SLAM systems are outlined, ranging from robotics and autonomous vehicles to augmented reality.

Backlight and dim space object detection based on a novel event camera

Background

For space object detection tasks, conventional optical cameras face various application challenges, including backlight issues and dim light conditions. As a novel optical camera, the event camera has the advantages of high temporal resolution and high dynamic range due to asynchronous output characteristics, which provides a new solution to the above challenges. However, the asynchronous output characteristic of event cameras makes them incompatible with conventional object detection methods designed for frame images.

Methods

Asynchronous convolutional memory network (ACMNet) for processing event camera data is proposed to solve the problem of backlight and dim space object detection. The key idea of ACMNet is to first characterize the asynchronous event streams with the Event Spike Tensor (EST) voxel grid through the exponential kernel function, then extract spatial features using a feed-forward feature extraction network, and aggregate temporal features using a proposed convolutional spatiotemporal memory module ConvLSTM, and finally, the end-to-end object detection using continuous event streams is realized.

Results

Comparison experiments among ACMNet and classical object detection methods are carried out on Event_DVS_space7, which is a large-scale space synthetic event dataset based on event cameras. The results show that the performance of ACMNet is superior to the others, and the mAP is improved by 12.7% while maintaining the processing speed. Moreover, event cameras still have a good performance in backlight and dim light conditions where conventional optical cameras fail. This research offers a novel possibility for detection under intricate lighting and motion conditions, emphasizing the superior benefits of event cameras in the realm of space object detection.

Cross-correlation image analysis for real-time single particle tracking

Accurately measuring the translations of objects between images is essential in many fields, including biology, medicine, chemistry, and physics. One important application is tracking one or more particles by measuring their apparent displacements in a series of images. Popular methods, such as the center of mass, often require idealized scenarios to reach the shot noise limit of particle tracking and, therefore, are not generally applicable to multiple image types. More general methods, such as maximum likelihood estimation, reliably approach the shot noise limit, but are too computationally intense for use in real-time applications. These limitations are significant, as real-time, shot-noise-limited particle tracking is of paramount importance for feedback control systems. To fill this gap, we introduce a new cross-correlation-based algorithm that approaches shot-noise-limited displacement detection and a graphics processing unit-based implementation for real-time image analysis of a single particle.

Efficient non-line-of-sight tracking with computational neuromorphic imaging

Non-line-of-sight (NLOS) sensing is an emerging technique that is capable of detecting objects hidden behind a wall, around corners, or behind other obstacles. However, NLOS tracking of moving objects is challenging due to signal redundancy and background interference. Here, we demonstrate computational neuromorphic imaging with an event camera for NLOS tracking, unaffected by the relay surface, which can efficiently obtain non-redundant information. We show how this sensor, which responds to changes in luminance within dynamic speckle fields, allows us to capture the most relevant events for direct motion estimation. The experimental results confirm that our method has superior performance in terms of efficiency, and accuracy, which greatly benefits from focusing on well-defined NLOS object tracking.

E-Calib: A Fast, Robust, and Accurate Calibration Toolbox for Event Cameras

Event cameras triggered a paradigm shift in the computer vision community delineated by their asynchronous nature, low latency, and high dynamic range. Calibration of event cameras is always essential to account for the sensor intrinsic parameters and for 3D perception. However, conventional image-based calibration techniques are not applicable due to the asynchronous, binary output of the sensor. The current standard for calibrating event cameras relies on either blinking patterns or event-based image reconstruction algorithms. These approaches are difficult to deploy in factory settings and are affected by noise and artifacts degrading the calibration performance. To bridge these limitations, we present E-Calib, a novel, fast, robust, and accurate calibration toolbox for event cameras utilizing the asymmetric circle grid, for its robustness to out-of-focus scenes. E-Calib introduces an efficient reweighted least squares (eRWLS) method for feature extraction of the calibration pattern circles with sub-pixel accuracy and robustness to noise. In addition, a modified hierarchical clustering algorithm is devised to detect the calibration grid apart from the background clutter. The proposed method is tested in a variety of rigorous experiments for different event camera models, on circle grids with different geometric properties, on varying calibration trajectories and speeds, and under challenging illumination conditions. The results show that our approach outperforms the state-of-the-art in detection success rate, reprojection error, and pose estimation accuracy.

EventLFM: event camera integrated Fourier light field microscopy for ultrafast 3D imaging

Ultrafast 3D imaging is indispensable for visualizing complex and dynamic biological processes. Conventional scanning-based techniques necessitate an inherent trade-off between acquisition speed and space-bandwidth product (SBP). Emerging single-shot 3D wide-field techniques offer a promising alternative but are bottlenecked by the synchronous readout constraints of conventional CMOS systems, thus restricting data throughput to maintain high SBP at limited frame rates. To address this, we introduce EventLFM, a straightforward and cost-effective system that overcomes these challenges by integrating an event camera with Fourier light field microscopy (LFM), a state-of-the-art single-shot 3D wide-field imaging technique. The event camera operates on a novel asynchronous readout architecture, thereby bypassing the frame rate limitations inherent to conventional CMOS systems. We further develop a simple and robust event-driven LFM reconstruction algorithm that can reliably reconstruct 3D dynamics from the unique spatiotemporal measurements captured by EventLFM. Experimental results demonstrate that EventLFM can robustly reconstruct fast-moving and rapidly blinking 3D fluorescent samples at kHz frame rates. Furthermore, we highlight EventLFM's capability for imaging of blinking neuronal signals in scattering mouse brain tissues and 3D tracking of GFP-labeled neurons in freely moving C. elegans. We believe that the combined ultrafast speed and large 3D SBP offered by EventLFM may open up new possibilities across many biomedical applications.

Real-World Spatial Synchronization of Event-CMOS Cameras through Deep Learning: A Novel CNN-DGCNN Approach

This paper presents a new deep-learning architecture designed to enhance the spatial synchronization between CMOS and event cameras by harnessing their complementary characteristics. While CMOS cameras produce high-quality imagery, they struggle in rapidly changing environments-a limitation that event cameras overcome due to their superior temporal resolution and motion clarity. However, effective integration of these two technologies relies on achieving precise spatial alignment, a challenge unaddressed by current algorithms. Our architecture leverages a dynamic graph convolutional neural network (DGCNN) to process event data directly, improving synchronization accuracy. We found that synchronization precision strongly correlates with the spatial concentration and density of events, with denser distributions yielding better alignment results. Our empirical results demonstrate that areas with denser event clusters enhance calibration accuracy, with calibration errors increasing in more uniformly distributed event scenarios. This research pioneers scene-based synchronization between CMOS and event cameras, paving the way for advancements in mixed-modality visual systems. The implications are significant for applications requiring detailed visual and temporal information, setting new directions for the future of visual perception technologies.

Convolutional neural network for improved event-based Shack-Hartmann wavefront reconstruction

Shack-Hartmann wavefront sensing is a technique for measuring wavefront aberrations, whose use in adaptive optics relies on fast position tracking of an array of spots. These sensors conventionally use frame-based cameras operating at a fixed sampling rate to report pixel intensities, even though only a fraction of the pixels have signal. Prior in-lab experiments have shown feasibility of event-based cameras for Shack-Hartmann wavefront sensing (SHWFS), asynchronously reporting the spot locations as log intensity changes at a microsecond time scale. In our work, we propose a convolutional neural network (CNN) called event-based wavefront network (EBWFNet) that achieves highly accurate estimation of the spot centroid position in real time. We developed a custom Shack-Hartmann wavefront sensing hardware with a common aperture for the synchronized frame- and event-based cameras so that spot centroid locations computed from the frame-based camera may be used to train/test the event-CNN-based centroid position estimation method in an unsupervised manner. Field testing with this hardware allows us to conclude that the proposed EBWFNet achieves sub-pixel accuracy in real-world scenarios with substantial improvement over the state-of-the-art event-based SHWFS. An ablation study reveals the impact of data processing, CNN components, and training cost function; and an unoptimized MATLAB implementation is shown to run faster than 800 Hz on a single GPU.

Microsaccade-inspired event camera for robotics

Neuromorphic vision sensors or event cameras have made the visual perception of extremely low reaction time possible, opening new avenues for high-dynamic robotics applications. These event cameras' output is dependent on both motion and texture. However, the event camera fails to capture object edges that are parallel to the camera motion. This is a problem intrinsic to the sensor and therefore challenging to solve algorithmically. Human vision deals with perceptual fading using the active mechanism of small involuntary eye movements, the most prominent ones called microsaccades. By moving the eyes constantly and slightly during fixation, microsaccades can substantially maintain texture stability and persistence. Inspired by microsaccades, we designed an event-based perception system capable of simultaneously maintaining low reaction time and stable texture. In this design, a rotating wedge prism was mounted in front of the aperture of an event camera to redirect light and trigger events. The geometrical optics of the rotating wedge prism allows for algorithmic compensation of the additional rotational motion, resulting in a stable texture appearance and high informational output independent of external motion. The hardware device and software solution are integrated into a system, which we call artificial microsaccade-enhanced event camera (AMI-EV). Benchmark comparisons validated the superior data quality of AMI-EV recordings in scenarios where both standard cameras and event cameras fail to deliver. Various real-world experiments demonstrated the potential of the system to facilitate robotics perception both for low-level and high-level vision tasks.

Spike-based dynamic computing with asynchronous sensing-computing neuromorphic chip

By mimicking the neurons and synapses of the human brain and employing spiking neural networks on neuromorphic chips, neuromorphic computing offers a promising energy-efficient machine intelligence. How to borrow high-level brain dynamic mechanisms to help neuromorphic computing achieve energy advantages is a fundamental issue. This work presents an application-oriented algorithm-software-hardware co-designed neuromorphic system for this issue. First, we design and fabricate an asynchronous chip called "Speck", a sensing-computing neuromorphic system on chip. With the low processor resting power of 0.42mW, Speck can satisfy the hardware requirements of dynamic computing: no-input consumes no energy. Second, we uncover the "dynamic imbalance" in spiking neural networks and develop an attention-based framework for achieving the algorithmic requirements of dynamic computing: varied inputs consume energy with large variance. Together, we demonstrate a neuromorphic system with real-time power as low as 0.70mW. This work exhibits the promising potentials of neuromorphic computing with its asynchronous event-driven, sparse, and dynamic nature.

Wearable robots for the real world need vision

To enhance wearable robots, understanding user intent and environmental perception with novel vision approaches is needed.

A Reconfigurable Bipolar Image Sensor for High-Efficiency Dynamic Vision Recognition

Dynamic vision perception and processing (DVPP) is in high demand by booming edge artificial intelligence. However, existing imaging systems suffer from low efficiency or low compatibility with advanced machine vision techniques. Here, we propose a reconfigurable bipolar image sensor (RBIS) for in-sensor DVPP based on a two-dimensional WSe2/GeSe heterostructure device. Owing to the gate-tunable and reversible built-in electric field, its photoresponse shows bipolarity as being positive or negative. High-efficiency DVPP incorporating front-end RBIS and back-end CNN is then demonstrated. It shows a high recognition accuracy of over 94.9% on the derived DVS128 data set and requires much fewer neural network parameters than that without RBIS. Moreover, we demonstrate an optimized device with a vertically stacked structure and a stable nonvolatile bipolarity, which enables more efficient DVPP hardware. Our work demonstrates the potential of fabricating DVPP devices with a simple structure, high efficiency, and outputs compatible with advanced algorithms.

Fully neuromorphic vision and control for autonomous drone flight

Biological sensing and processing is asynchronous and sparse, leading to low-latency and energy-efficient perception and action. In robotics, neuromorphic hardware for event-based vision and spiking neural networks promises to exhibit similar characteristics. However, robotic implementations have been limited to basic tasks with low-dimensional sensory inputs and motor actions because of the restricted network size in current embedded neuromorphic processors and the difficulties of training spiking neural networks. Here, we present a fully neuromorphic vision-to-control pipeline for controlling a flying drone. Specifically, we trained a spiking neural network that accepts raw event-based camera data and outputs low-level control actions for performing autonomous vision-based flight. The vision part of the network, consisting of five layers and 28,800 neurons, maps incoming raw events to ego-motion estimates and was trained with self-supervised learning on real event data. The control part consists of a single decoding layer and was learned with an evolutionary algorithm in a drone simulator. Robotic experiments show a successful sim-to-real transfer of the fully learned neuromorphic pipeline. The drone could accurately control its ego-motion, allowing for hovering, landing, and maneuvering sideways-even while yawing at the same time. The neuromorphic pipeline runs on board on Intel's Loihi neuromorphic processor with an execution frequency of 200 hertz, consuming 0.94 watt of idle power and a mere additional 7 to 12 milliwatts when running the network. These results illustrate the potential of neuromorphic sensing and processing for enabling insect-sized intelligent robots.

Detecting Line Segments in Motion-Blurred Images With Events

Making line segment detectors more reliable under motion blurs is one of the most important challenges for practical applications, such as visual SLAM and 3D line mapping. Existing line segment detection methods face severe performance degradation for accurately detecting and locating line segments when motion blur occurs. While event data shows strong complementary characteristics to images for minimal blur and edge awareness at high-temporal resolution, potentially beneficial for reliable line segment recognition. To robustly detect line segments over motion blurs, we propose to leverage the complementary information of images and events. Specifically, we first design a general frame-event feature fusion network to extract and fuse the detailed image textures and low-latency event edges, which consists of a channel-attention-based shallow fusion module and a self-attention-based dual hourglass module. We then utilize the state-of-the-art wireframe parsing networks to detect line segments on the fused feature map. Moreover, due to the lack of line segment detection datasets with pairwise motion-blurred images and events, we contribute two datasets, i.e., synthetic FE-Wireframe and realistic FE-Blurframe, for network training and evaluation. Extensive analyses on the component configurations demonstrate the design effectiveness of our fusion network. When compared to the state-of-the-arts, the proposed approach achieves the highest detection accuracy while maintaining comparable real-time performance. In addition to being robust to motion blur, our method also exhibits superior performance for line detection under high dynamic range scenes.

Swift-Eye: Towards Anti-blink Pupil Tracking for Precise and Robust High-Frequency Near-Eye Movement Analysis with Event Cameras

Eye tracking has shown great promise in many scientific fields and daily applications, ranging from the early detection of mental health disorders to foveated rendering in virtual reality (VR). These applications all call for a robust system for high-frequency near-eye movement sensing and analysis in high precision, which cannot be guaranteed by the existing eye tracking solutions with CCD/CMOS cameras. To bridge the gap, in this paper, we propose Swift-Eye, an offline precise and robust pupil estimation and tracking framework to support high-frequency near-eye movement analysis, especially when the pupil region is partially occluded. Swift-Eye is built upon the emerging event cameras to capture the high-speed movement of eyes in high temporal resolution. Then, a series of bespoke components are designed to generate high-quality near-eye movement video at a high frame rate over kilohertz and deal with the occlusion over the pupil caused by involuntary eye blinks. According to our extensive evaluations on EV-Eye, a large-scale public dataset for eye tracking using event cameras, Swift-Eye shows high robustness against significant occlusion. It can improve the IoU and F1-score of the pupil estimation by 20% and 12.5% respectively, compared with the second-best competing approach, when over 80% of the pupil region is occluded by the eyelid. Lastly, it provides continuous and smooth traces of pupils in extremely high temporal resolution and can support high-frequency eye movement analysis and a number of potential applications, such as mental health diagnosis, behaviour-brain association, etc. The implementation details and source codes can be found at https://github.com/ztysdu/Swift-Eye.

Low-latency automotive vision with event cameras

The computer vision algorithms used currently in advanced driver assistance systems rely on image-based RGB cameras, leading to a critical bandwidth-latency trade-off for delivering safe driving experiences. To address this, event cameras have emerged as alternative vision sensors. Event cameras measure the changes in intensity asynchronously, offering high temporal resolution and sparsity, markedly reducing bandwidth and latency requirements1. Despite these advantages, event-camera-based algorithms are either highly efficient but lag behind image-based ones in terms of accuracy or sacrifice the sparsity and efficiency of events to achieve comparable results. To overcome this, here we propose a hybrid event- and frame-based object detector that preserves the advantages of each modality and thus does not suffer from this trade-off. Our method exploits the high temporal resolution and sparsity of events and the rich but low temporal resolution information in standard images to generate efficient, high-rate object detections, reducing perceptual and computational latency. We show that the use of a 20 frames per second (fps) RGB camera plus an event camera can achieve the same latency as a 5,000-fps camera with the bandwidth of a 45-fps camera without compromising accuracy. Our approach paves the way for efficient and robust perception in edge-case scenarios by uncovering the potential of event cameras2.

Motion measurements of explosive shock waves based on an event camera

Shock wave measurement is vital in assessing explosive power and designing warheads. To obtain satisfactory observation data of explosive shock waves, it is preferable for optical sensors to possess high-dynamic range and high-time resolution capabilities. In this paper, the event camera is first employed to observe explosive shock waves, leveraging its high dynamic range and low latency. A comprehensive procedure is devised to measure the motion parameters of shock waves accurately. Firstly, the plane lines-based calibration method is proposed to compute the calibration parameters of the event camera, which utilizes the edge-sensitive characteristic of the event camera. Then, the fitted ellipse parameters of the shock wave are estimated based on the concise event data, which are gained by utilizing the characteristics of the event triggering and shock waves' morphology. Finally, the geometric relationship between the ellipse parameters and the radius of the shock wave is derived, and the motion parameters of the shock wave are estimated. To verify the performance of our method, we compare our measurement results in the TNT explosion test with the pressure sensor results and empirical formula prediction. The relative measurement error compared to pressure sensors is the lowest at 0.33% and the highest at 7.58%. The experimental results verify the rationality and effectiveness of our methods.

Surrogate gradient learning in spiking networks trained on event-based cytometry dataset

Spiking neural networks (SNNs) are bio-inspired neural networks that - to an extent - mimic the workings of our brains. In a similar fashion, event-based vision sensors try to replicate a biological eye as closely as possible. In this work, we integrate both technologies for the purpose of classifying micro-particles in the context of label-free flow cytometry. We follow up on our previous work in which we used simple logistic regression with binary labels. Although this model was able to achieve an accuracy of over 98%, our goal is to utilize the system for a wider variety of cells, some of which may have less noticeable morphological variations. Therefore, a more advanced machine learning model like the SNNs discussed here would be required. This comes with the challenge of training such networks, since they typically suffer from vanishing gradients. We effectively apply the surrogate gradient method to overcome this issue achieving over 99% classification accuracy on test data for a four-class problem. Finally, rather than treating the neural network as a black box, we explore the dynamics inside the network and make use of that to enhance its accuracy and sparsity.

Memristive tonotopic mapping with volatile resistive switching memory devices

To reach the energy efficiency and the computing capability of biological neural networks, novel hardware systems and paradigms are required where the information needs to be processed in both spatial and temporal domains. Resistive switching memory (RRAM) devices appear as key enablers for the implementation of large-scale neuromorphic computing systems with high energy efficiency and extended scalability. Demonstrating a full set of spatiotemporal primitives with RRAM-based circuits remains an open challenge. By taking inspiration from the neurobiological processes in the human auditory systems, we develop neuromorphic circuits for memristive tonotopic mapping via volatile RRAM devices. Based on a generalized stochastic device-level approach, we demonstrate the main features of signal processing of cochlea, namely logarithmic integration and tonotopic mapping of signals. We also show that our tonotopic classification is suitable for speech recognition. These results support memristive devices for physical processing of temporal signals, thus paving the way for energy efficient, high density neuromorphic systems.

Adaptive Unsupervised Learning-Based 3D Spatiotemporal Filter for Event-Driven Cameras

In the evolving landscape of robotics and visual navigation, event cameras have gained important traction, notably for their exceptional dynamic range, efficient power consumption, and low latency. Despite these advantages, conventional processing methods oversimplify the data into 2 dimensions, neglecting critical temporal information. To overcome this limitation, we propose a novel method that treats events as 3D time-discrete signals. Drawing inspiration from the intricate biological filtering systems inherent to the human visual apparatus, we have developed a 3D spatiotemporal filter based on unsupervised machine learning algorithm. This filter effectively reduces noise levels and performs data size reduction, with its parameters being dynamically adjusted based on population activity. This ensures adaptability and precision under various conditions, like changes in motion velocity and ambient lighting. In our novel validation approach, we first identify the noise type and determine its power spectral density in the event stream. We then apply a one-dimensional discrete fast Fourier transform to assess the filtered event data within the frequency domain, ensuring that the targeted noise frequencies are adequately reduced. Our research also delved into the impact of indoor lighting on event stream noise. Remarkably, our method led to a 37% decrease in the data point cloud, improving data quality in diverse outdoor settings.

Optimizing event-based neural networks on digital neuromorphic architecture: a comprehensive design space exploration

Neuromorphic processors promise low-latency and energy-efficient processing by adopting novel brain-inspired design methodologies. Yet, current neuromorphic solutions still struggle to rival conventional deep learning accelerators' performance and area efficiency in practical applications. Event-driven data-flow processing and near/in-memory computing are the two dominant design trends of neuromorphic processors. However, there remain challenges in reducing the overhead of event-driven processing and increasing the mapping efficiency of near/in-memory computing, which directly impacts the performance and area efficiency. In this work, we discuss these challenges and present our exploration of optimizing event-based neural network inference on SENECA, a scalable and flexible neuromorphic architecture. To address the overhead of event-driven processing, we perform comprehensive design space exploration and propose spike-grouping to reduce the total energy and latency. Furthermore, we introduce the event-driven depth-first convolution to increase area efficiency and latency in convolutional neural networks (CNNs) on the neuromorphic processor. We benchmarked our optimized solution on keyword spotting, sensor fusion, digit recognition and high resolution object detection tasks. Compared with other state-of-the-art large-scale neuromorphic processors, our proposed optimizations result in a 6× to 300× improvement in energy efficiency, a 3× to 15× improvement in latency, and a 3× to 100× improvement in area efficiency. Our optimizations for event-based neural networks can be potentially generalized to a wide range of event-based neuromorphic processors.

SLoN: a spiking looming perception network exploiting neural encoding and processing in ON/OFF channels

Looming perception, the ability to sense approaching objects, is crucial for the survival of humans and animals. After hundreds of millions of years of evolutionary development, biological entities have evolved efficient and robust looming perception visual systems. However, current artificial vision systems fall short of such capabilities. In this study, we propose a novel spiking neural network for looming perception that mimics biological vision to communicate motion information through action potentials or spikes, providing a more realistic approach than previous artificial neural networks based on sum-then-activate operations. The proposed spiking looming perception network (SLoN) comprises three core components. Neural encoding, known as phase coding, transforms video signals into spike trains, introducing the concept of phase delay to depict the spatial-temporal competition between phasic excitatory and inhibitory signals shaping looming selectivity. To align with biological substrates where visual signals are bifurcated into parallel ON/OFF channels encoding brightness increments and decrements separately to achieve specific selectivity to ON/OFF-contrast stimuli, we implement eccentric down-sampling at the entrance of ON/OFF channels, mimicking the foveal region of the mammalian receptive field with higher acuity to motion, computationally modeled with a leaky integrate-and-fire (LIF) neuronal network. The SLoN model is deliberately tested under various visual collision scenarios, ranging from synthetic to real-world stimuli. A notable achievement is that the SLoN selectively spikes for looming features concealed in visual streams against other categories of movements, including translating, receding, grating, and near misses, demonstrating robust selectivity in line with biological principles. Additionally, the efficacy of the ON/OFF channels, the phase coding with delay, and the eccentric visual processing are further investigated to demonstrate their effectiveness in looming perception. The cornerstone of this study rests upon showcasing a new paradigm for looming perception that is more biologically plausible in light of biological motion perception.

Colorful image reconstruction from neuromorphic event cameras with biologically inspired deep color fusion neural networks

Neuromorphic event-based cameras communicate transients in luminance instead of frames, providing visual information with a fine temporal resolution, high dynamic range and high signal-to-noise ratio. Enriching event data with color information allows for the reconstruction of colorful frame-like intensity maps, supporting improved performance and visually appealing results in various computer vision tasks. In this work, we simulated a biologically inspired color fusion system featuring a three-stage convolutional neural network for reconstructing color intensity maps from event data and sparse color cues. While current approaches for color fusion use full RGB frames in high resolution, our design uses event data and low-spatial and tonal-resolution quantized color cues, providing a high-performing small model for efficient colorful image reconstruction. The proposed model outperforms existing coloring schemes in terms of SSIM, LPIPS, PSNR, and CIEDE2000 metrics. We demonstrate that auxiliary limited color information can be used in conjunction with event data to successfully reconstruct both color and intensity frames, paving the way for more efficient hardware designs.

Neuromorphic Camera Denoising Using Graph Neural Network-Driven Transformers

Neuromorphic vision is a bio-inspired technology that has triggered a paradigm shift in the computer vision community and is serving as a key enabler for a wide range of applications. This technology has offered significant advantages, including reduced power consumption, reduced processing needs, and communication speedups. However, neuromorphic cameras suffer from significant amounts of measurement noise. This noise deteriorates the performance of neuromorphic event-based perception and navigation algorithms. In this article, we propose a novel noise filtration algorithm to eliminate events that do not represent real log-intensity variations in the observed scene. We employ a graph neural network (GNN)-driven transformer algorithm, called GNN-Transformer, to classify every active event pixel in the raw stream into real log-intensity variation or noise. Within the GNN, a message-passing framework, referred to as EventConv, is carried out to reflect the spatiotemporal correlation among the events while preserving their asynchronous nature. We also introduce the known-object ground-truth labeling (KoGTL) approach for generating approximate ground-truth labels of event streams under various illumination conditions. KoGTL is used to generate labeled datasets, from experiments recorded in challenging lighting conditions, including moon light. These datasets are used to train and extensively test our proposed algorithm. When tested on unseen datasets, the proposed algorithm outperforms state-of-the-art methods by at least 8.8% in terms of filtration accuracy. Additional tests are also conducted on publicly available datasets (ETH Zürich Color-DAVIS346 datasets) to demonstrate the generalization capabilities of the proposed algorithm in the presence of illumination variations and different motion dynamics. Compared to state-of-the-art solutions, qualitative results verified the superior capability of the proposed algorithm to eliminate noise while preserving meaningful events in the scene.

Faces in Event Streams (FES): An Annotated Face Dataset for Event Cameras

The use of event-based cameras in computer vision is a growing research direction. However, despite the existing research on face detection using the event camera, a substantial gap persists in the availability of a large dataset featuring annotations for faces and facial landmarks on event streams, thus hampering the development of applications in this direction. In this work, we address this issue by publishing the first large and varied dataset (Faces in Event Streams) with a duration of 689 min for face and facial landmark detection in direct event-based camera outputs. In addition, this article presents 12 models trained on our dataset to predict bounding box and facial landmark coordinates with an mAP50 score of more than 90%. We also performed a demonstration of real-time detection with an event-based camera using our models.

Long- and Short-Term Memory Model of Cotton Price Index Volatility Risk Based on Explainable Artificial Intelligence

Market uncertainty greatly interferes with the decisions and plans of market participants, thus increasing the risk of decision-making, leading to compromised interests of decision-makers. Cotton price index (hereinafter referred to as cotton price) volatility is highly noisy, nonlinear, and stochastic and is susceptible to supply and demand, climate, substitutes, and other policy factors, which are subject to large uncertainties. To reduce decision risk and provide decision support for policymakers, this article integrates 13 factors affecting cotton price index volatility based on existing research and further divides them into transaction data and interaction data. A long- and short-term memory (LSTM) model is constructed, and a comparison experiment is implemented to analyze the cotton price index volatility. To make the constructed model explainable, we use explainable artificial intelligence (XAI) techniques to perform statistical analysis of the input features. The experimental results show that the LSTM model can accurately analyze the cotton price index fluctuation trend but cannot accurately predict the actual price of cotton; the transaction data plus interaction data are more sensitive than the transaction data in analyzing the cotton price fluctuation trend and can have a positive effect on the cotton price fluctuation analysis. This study can accurately reflect the fluctuation trend of the cotton market, provide reference to the state, enterprises, and cotton farmers for decision-making, and reduce the risk caused by frequent fluctuation of cotton prices. The analysis of the model using XAI techniques builds the confidence of decision-makers in the model.

Special Issue-Biosensors and Neuroscience: Is Biosensors Engineering Ready to Embrace Design Principles from Neuroscience?

In partnership with the Air Force Office of Scientific Research (AFOSR), the National Science Foundation's (NSF) Emerging Frontiers and Multidisciplinary Activities (EFMA) office of the Directorate for Engineering (ENG) launched an Emerging Frontiers in Research and Innovation (EFRI) topic for the fiscal years FY22 and FY23 entitled "Brain-inspired Dynamics for Engineering Energy-Efficient Circuits and Artificial Intelligence" (BRAID) [...].

Mathematically Improved XGBoost Algorithm for Truck Hoisting Detection in Container Unloading

Truck hoisting detection constitutes a key focus in port security, for which no optimal resolution has been identified. To address the issues of high costs, susceptibility to weather conditions, and low accuracy in conventional methods for truck hoisting detection, a non-intrusive detection approach is proposed in this paper. The proposed approach utilizes a mathematical model and an extreme gradient boosting (XGBoost) model. Electrical signals, including voltage and current, collected by Hall sensors are processed by the mathematical model, which augments their physical information. Subsequently, the dataset filtered by the mathematical model is used to train the XGBoost model, enabling the XGBoost model to effectively identify abnormal hoists. Improvements were observed in the performance of the XGBoost model as utilized in this paper. Finally, experiments were conducted at several stations. The overall false positive rate did not exceed 0.7% and no false negatives occurred in the experiments. The experimental results demonstrated the excellent performance of the proposed approach, which can reduce the costs and improve the accuracy of detection in container hoisting.

Moving event detection from LiDAR point streams

In dynamic environments, robots require instantaneous detection of moving events with microseconds of latency. This task, known as moving event detection, is typically achieved using event cameras. While light detection and ranging (LiDAR) sensors are essential for robots due to their dense and accurate depth measurements, their use in event detection has not been thoroughly explored. Current approaches involve accumulating LiDAR points into frames and detecting object-level motions, resulting in a latency of tens to hundreds of milliseconds. We present a different approach called M-detector, which determines if a point is moving immediately after its arrival, resulting in a point-by-point detection with a latency of just several microseconds. M-detector is designed based on occlusion principles and can be used in different environments with various types of LiDAR sensors. Our experiments demonstrate the effectiveness of M-detector on various datasets and applications, showcasing its superior accuracy, computational efficiency, detection latency, and generalization ability.

An FPGA implementation of Bayesian inference with spiking neural networks

Spiking neural networks (SNNs), as brain-inspired neural network models based on spikes, have the advantage of processing information with low complexity and efficient energy consumption. Currently, there is a growing trend to design hardware accelerators for dedicated SNNs to overcome the limitation of running under the traditional von Neumann architecture. Probabilistic sampling is an effective modeling approach for implementing SNNs to simulate the brain to achieve Bayesian inference. However, sampling consumes considerable time. It is highly demanding for specific hardware implementation of SNN sampling models to accelerate inference operations. Hereby, we design a hardware accelerator based on FPGA to speed up the execution of SNN algorithms by parallelization. We use streaming pipelining and array partitioning operations to achieve model operation acceleration with the least possible resource consumption, and combine the Python productivity for Zynq (PYNQ) framework to implement the model migration to the FPGA while increasing the speed of model operations. We verify the functionality and performance of the hardware architecture on the Xilinx Zynq ZCU104. The experimental results show that the hardware accelerator of the SNN sampling model proposed can significantly improve the computing speed while ensuring the accuracy of inference. In addition, Bayesian inference for spiking neural networks through the PYNQ framework can fully optimize the high performance and low power consumption of FPGAs in embedded applications. Taken together, our proposed FPGA implementation of Bayesian inference with SNNs has great potential for a wide range of applications, it can be ideal for implementing complex probabilistic model inference in embedded systems.

Chip-In-Loop SNN Proxy Learning: a new method for efficient training of spiking neural networks

The primary approaches used to train spiking neural networks (SNNs) involve either training artificial neural networks (ANNs) first and then transforming them into SNNs, or directly training SNNs using surrogate gradient techniques. Nevertheless, both of these methods encounter a shared challenge: they rely on frame-based methodologies, where asynchronous events are gathered into synchronous frames for computation. This strays from the authentic asynchronous, event-driven nature of SNNs, resulting in notable performance degradation when deploying the trained models on SNN simulators or hardware chips for real-time asynchronous computation. To eliminate this performance degradation, we propose a hardware-based SNN proxy learning method that is called Chip-In-Loop SNN Proxy Learning (CIL-SPL). This approach effectively eliminates the performance degradation caused by the mismatch between synchronous and asynchronous computations. To demonstrate the effectiveness of our method, we trained models using public datasets such as N-MNIST and tested them on the SNN simulator or hardware chip, comparing our results to those classical training methods.

Widefield Diamond Quantum Sensing with Neuromorphic Vision Sensors

Despite increasing interest in developing ultrasensitive widefield diamond magnetometry for various applications, achieving high temporal resolution and sensitivity simultaneously remains a key challenge. This is largely due to the transfer and processing of massive amounts of data from the frame-based sensor to capture the widefield fluorescence intensity of spin defects in diamonds. In this study, a neuromorphic vision sensor to encode the changes of fluorescence intensity into spikes in the optically detected magnetic resonance (ODMR) measurements is adopted, closely resembling the operation of the human vision system, which leads to highly compressed data volume and reduced latency. It also results in a vast dynamic range, high temporal resolution, and exceptional signal-to-background ratio. After a thorough theoretical evaluation, the experiment with an off-the-shelf event camera demonstrated a 13× improvement in temporal resolution with comparable precision of detecting ODMR resonance frequencies compared with the state-of-the-art highly specialized frame-based approach. It is successfully deploy this technology in monitoring dynamically modulated laser heating of gold nanoparticles coated on a diamond surface, a recognizably difficult task using existing approaches. The current development provides new insights for high-precision and low-latency widefield quantum sensing, with possibilities for integration with emerging memory devices to realize more intelligent quantum sensors.

Fast 3D reconstruction via event-based structured light with spatio-temporal coding

Event-based structured light (SL) systems leverage bio-inspired event cameras, which are renowned for their low latency and high dynamics, to drive progress in high-speed structured light systems. However, existing event-based structured light methods concentrate on the independent construction of either time-domain or space-domain features for stereo matching, ignoring the spatio-temporal consistency towards depth. In this work, we build an event-based SL system that consists of a laser point projector and an event camera, and we devise a spatial-temporal coding strategy that realizes depth encoding in dual domains through a single shot. To exploit the spatio-temporal synergy, we further present STEM, a novel Spatio-Temporal Enhanced Matching approach for 3D reconstruction. STEM is comprised of two parts, the spatio-temporal enhancing (STE) algorithm and the spatio-temporal matching (STM) algorithm. Specifically, STE integrates the dual-domain information to increase the saliency of the temporal coding, providing a more robust basis for matching. STM is a stereo matching algorithm explicitly tailored to the unique characteristics of event data modality, which computes the disparity via a meticulously designed hybrid cost function. Experimental results demonstrate the superior performance of our proposed method, achieving a reconstruction rate of 16 fps and a low root mean square error of 0.56 mm at a distance of 0.72 m.

Velocity and Color Estimation Using Event-Based Clustering

Event-based clustering provides a low-power embedded solution for low-level feature extraction in a scene. The algorithm utilizes the non-uniform sampling capability of event-based image sensors to measure local intensity variations within a scene. Consequently, the clustering algorithm forms similar event groups while simultaneously estimating their attributes. This work proposes taking advantage of additional event information in order to provide new attributes for further processing. We elaborate on the estimation of the object velocity using the mean motion of the cluster. Next, we are examining a novel form of events, which includes intensity measurement of the color at the concerned pixel. These events may be processed to estimate the rough color of a cluster, or the color distribution in a cluster. Lastly, this paper presents some applications that utilize these features. The resulting algorithms are applied and exercised thanks to a custom event-based simulator, which generates videos of outdoor scenes. The velocity estimation methods provide satisfactory results with a trade-off between accuracy and convergence speed. Regarding color estimation, the luminance estimation is challenging in the test cases, while the chrominance is precisely estimated. The estimated quantities are adequate for accurately classifying objects into predefined categories.

NeuroZoom: Denoising and Super Resolving Neuromorphic Events and Spikes

Neuromorphic cameras are emerging imaging technology that has advantages over conventional imaging sensors in several aspects including dynamic range, sensing latency, and power consumption. However, the signal-to-noise level and the spatial resolution still fall behind the state of conventional imaging sensors. In this article, we address the denoising and super-resolution problem for modern neuromorphic cameras. We employ 3D U-Net as the backbone neural architecture for such a task. The networks are trained and tested on two types of neuromorphic cameras: a dynamic vision sensor and a spike camera. Their pixels generate signals asynchronously, the former is based on perceived light changes and the latter is based on accumulated light intensity. To collect the datasets for training such networks, we design a display-camera system to record high frame-rate videos at multiple resolutions, providing supervision for denoising and super-resolution. The networks are trained in a noise-to-noise fashion, where the two ends of the network are unfiltered noisy data. The output of the networks has been tested for downstream applications including event-based visual object tracking and image reconstruction. Experimental results demonstrate the effectiveness of improving the quality of neuromorphic events and spikes, and the corresponding improvement to downstream applications with state-of-the-art performance.

STCA-SNN: self-attention-based temporal-channel joint attention for spiking neural networks

Spiking Neural Networks (SNNs) have shown great promise in processing spatio-temporal information compared to Artificial Neural Networks (ANNs). However, there remains a performance gap between SNNs and ANNs, which impedes the practical application of SNNs. With intrinsic event-triggered property and temporal dynamics, SNNs have the potential to effectively extract spatio-temporal features from event streams. To leverage the temporal potential of SNNs, we propose a self-attention-based temporal-channel joint attention SNN (STCA-SNN) with end-to-end training, which infers attention weights along both temporal and channel dimensions concurrently. It models global temporal and channel information correlations with self-attention, enabling the network to learn 'what' and 'when' to attend simultaneously. Our experimental results show that STCA-SNNs achieve better performance on N-MNIST (99.67%), CIFAR10-DVS (81.6%), and N-Caltech 101 (80.88%) compared with the state-of-the-art SNNs. Meanwhile, our ablation study demonstrates that STCA-SNNs improve the accuracy of event stream classification tasks.

Distractor-Aware Event-Based Tracking

Event cameras, or dynamic vision sensors, have recently achieved success from fundamental vision tasks to high-level vision researches. Due to its ability to asynchronously capture light intensity changes, event camera has an inherent advantage to capture moving objects in challenging scenarios including objects under low light, high dynamic range, or fast moving objects. Thus event camera are natural for visual object tracking. However, the current event-based trackers derived from RGB trackers simply modify the input images to event frames and still follow conventional tracking pipeline that mainly focus on object texture for target distinction. As a result, the trackers may not be robust dealing with challenging scenarios such as moving cameras and cluttered foreground. In this paper, we propose a distractor-aware event-based tracker that introduces transformer modules into Siamese network architecture (named DANet). Specifically, our model is mainly composed of a motion-aware network and a target-aware network, which simultaneously exploits both motion cues and object contours from event data, so as to discover motion objects and identify the target object by removing dynamic distractors. Our DANet can be trained in an end-to-end manner without any post-processing and can run at over 80 FPS on a single V100. We conduct comprehensive experiments on two large event tracking datasets to validate the proposed model. We demonstrate that our tracker has superior performance against the state-of-the-art trackers in terms of both accuracy and efficiency.