Learning the 3-D structure of objects from 2-D views depends on shape, not format

The analysis of the structural parameter influences on measurement errors in a binocular 3D reconstruction system: a portable 3D system

This study used an analytical model to investigate the factors that affect the reconstruction accuracy composed of the baseline length, lens focal length, the angle between the optical axis and baseline, and the field of the view angle. Firstly, the theoretical expressions of the above factors and measurement errors are derived based on the binocular three-dimensional reconstruction model. Then, the structural parameters' impact on the error propagation coefficient is analyzed and simulated using MATLAB software. The results show that structural parameters significantly impact the error propagation coefficient, and the reasonable range of structural parameters is pointed out. When the angle between the optical axis of the binocular camera and the baseline is between 30° and 55°, the ratio of the baseline length to the focal length can be reasonably reduced. In addition, using the field angle of the view that does not exceed 20° could reduce the error propagation coefficient. While the angle between the binocular optical axis and the baseline is between 40° and 50°, the reconstruction result has the highest accuracy, changing the angle out of this range will lead to an increase in the reconstruction error. The angle between the binocular optical axis and the baseline changes 30° through 60° leads to the error propagation coefficient being in a lower range. Finally, experimental verification and simulation results show that selecting reasonable structural parameters could significantly reduce measurement errors. This study proposes a model that constructs a binocular three-dimensional reconstruction system with high precision. A portable three-dimensional reconstruction system is built in the article.

Standardised images of novel objects created with generative adversarial networks

An enduring question in cognitive science is how perceptually novel objects are processed. Addressing this issue has been limited by the absence of a standardised set of object-like stimuli that appear realistic, but cannot possibly have been previously encountered. To this end, we created a dataset, at the core of which are images of 400 perceptually novel objects. These stimuli were created using Generative Adversarial Networks that integrated features of everyday stimuli to produce a set of synthetic objects that appear entirely plausible, yet do not in fact exist. We curated an accompanying dataset of 400 familiar stimuli, which were matched in terms of size, contrast, luminance, and colourfulness. For each object, we quantified their key visual properties (edge density, entropy, symmetry, complexity, and spectral signatures). We also confirmed that adult observers (N = 390) perceive the novel objects to be less familiar, yet similarly engaging, relative to the familiar objects. This dataset serves as an open resource to facilitate future studies on visual perception.

Stereopsis provides a constant feed to visual shape representation

The contribution of stereopsis in human visual shape perception was examined using stimuli with either null, normal, or reversed binocular disparity in an old/new object recognition task. The highest levels of recognition performance were observed with null and normal binocular disparity displays, which did not differ. However, reversed disparity led to significantly worse performance than either of the other display conditions. This indicates that stereopsis provides a continuous input to the mechanisms involved in shape perception.

Combined Neural Tuning in Human Ventral Temporal Cortex Resolves the Perceptual Ambiguity of Morphed 2D Images

We have an amazing ability to categorize objects in the world around us. Nevertheless, how cortical regions in human ventral temporal cortex (VTC), which is critical for categorization, support this behavioral ability, is largely unknown. Here, we examined the relationship between neural responses and behavioral performance during the categorization of morphed silhouettes of faces and hands, which are animate categories processed in cortically adjacent regions in VTC. Our results reveal that the combination of neural responses from VTC face- and body-selective regions more accurately explains behavioral categorization than neural responses from either region alone. Furthermore, we built a model that predicts a person's behavioral performance using estimated parameters of brain-behavior relationships from a different group of people. Moreover, we show that this brain-behavior model generalizes to adjacent face- and body-selective regions in lateral occipitotemporal cortex. Thus, while face- and body-selective regions are located within functionally distinct domain-specific networks, cortically adjacent regions from both networks likely integrate neural responses to resolve competing and perceptually ambiguous information from both categories.

The functional neuroanatomy of face perception: from brain measurements to deep neural networks

A central goal in neuroscience is to understand how processing within the ventral visual stream enables rapid and robust perception and recognition. Recent neuroscientific discoveries have significantly advanced understanding of the function, structure and computations along the ventral visual stream that serve as the infrastructure supporting this behaviour. In parallel, significant advances in computational models, such as hierarchical deep neural networks (DNNs), have brought machine performance to a level that is commensurate with human performance. Here, we propose a new framework using the ventral face network as a model system to illustrate how increasing the neural accuracy of present DNNs may allow researchers to test the computational benefits of the functional architecture of the human brain. Thus, the review (i) considers specific neural implementational features of the ventral face network, (ii) describes similarities and differences between the functional architecture of the brain and DNNs, and (iii) provides a hypothesis for the computational value of implementational features within the brain that may improve DNN performance. Importantly, this new framework promotes the incorporation of neuroscientific findings into DNNs in order to test the computational benefits of fundamental organizational features of the visual system.