Learning to recognize three-dimensional objects

Cost-sensitive boosting

A novel framework is proposed for the design of cost-sensitive boosting algorithms. The framework is based on the identification of two necessary conditions for optimal cost-sensitive learning that 1) expected losses must be minimized by optimal cost-sensitive decision rules and 2) empirical loss minimization must emphasize the neighborhood of the target cost-sensitive boundary. It is shown that these conditions enable the derivation of cost-sensitive losses that can be minimized by gradient descent, in the functional space of convex combinations of weak learners, to produce novel boosting algorithms. The proposed framework is applied to the derivation of cost-sensitive extensions of AdaBoost, RealBoost, and LogitBoost. Experimental evidence, with a synthetic problem, standard data sets, and the computer vision problems of face and car detection, is presented in support of the cost-sensitive optimality of the new algorithms. Their performance is also compared to those of various previous cost-sensitive boosting proposals, as well as the popular combination of large-margin classifiers and probability calibration. Cost-sensitive boosting is shown to consistently outperform all other methods.

A Bayesian model for efficient visual search and recognition

Humans employ interacting bottom-up and top-down processes to significantly speed up search and recognition of particular targets. We describe a new model of attention guidance for efficient and scalable first-stage search and recognition with many objects (117,174 images of 1147 objects were tested, and 40 satellite images). Performance for recognition is on par or better than SIFT and HMAX, while being, respectively, 1500 and 279 times faster. The model is also used for top-down guided search, finding a desired object in a 5x5 search array within four attempts, and improving performance for finding houses in satellite images.

Learning representations for object classification using multi-stage optimal component analysis

Learning data representations is a fundamental challenge in modeling neural processes and plays an important role in applications such as object recognition. Optimal component analysis (OCA) formulates the problem in the framework of optimization on a Grassmann manifold and a stochastic gradient method is used to estimate the optimal basis. OCA has been successfully applied to image classification problems arising in a variety of contexts. However, as the search space is typically very high dimensional, OCA optimization often requires expensive computational cost. In multi-stage OCA, we first hierarchically project the data onto several low-dimensional subspaces using standard techniques, then OCA learning is performed hierarchically from the lowest to the highest levels to learn about a subspace that is optimal for data discrimination based on the K-nearest neighbor classifier. One of the main advantages of multi-stage OCA lies in the fact that it greatly improves the computational efficiency of the OCA learning algorithm without sacrificing the recognition performance, thus enhancing its applicability to practical problems. In addition to the nearest neighbor classifier, we illustrate the effectiveness of the learned representations on object classification used in conjunction with classifiers such as neural networks and support vector machines.

Object recognition by artificial cortical maps

Object recognition is one of the most important functions of the human visual system, yet one of the least understood, this despite the fact that vision is certainly the most studied function of the brain. We understand relatively well how several processes in the cortical visual areas that support recognition capabilities take place, such as orientation discrimination and color constancy. This paper proposes a model of the development of object recognition capability, based on two main theoretical principles. The first is that recognition does not imply any sort of geometrical reconstruction, it is instead fully driven by the two dimensional view captured by the retina. The second assumption is that all the processing functions involved in recognition are not genetically determined or hardwired in neural circuits, but are the result of interactions between epigenetic influences and basic neural plasticity mechanisms. The model is organized in modules roughly related to the main visual biological areas, and is implemented mainly using the LISSOM architecture, a recent neural self-organizing map model that simulates the effects of intercortical lateral connections. This paper shows how recognition capabilities, similar to those found in brain ventral visual areas, can develop spontaneously by exposure to natural images in an artificial cortical model.

Distinct multicolored region descriptors for object recognition

The problem of object recognition has been considered here. Color descriptions from distinct regions covering multiple segments are considered for object representation. Distinct multicolored regions are detected using edge maps and clustering. Performance of the proposed methodologies has been evaluated on three data sets and the results are found to be better than existing methods when a small number of training views is considered.

Sparse representation for coarse and fine object recognition

This paper offers a sparse, multiscale representation of objects. It captures the object appearance by selection from a very large dictionary of Gaussian differential basis functions. The learning procedure results from the matching pursuit algorithm, while the recognition is based on polynomial approximation to the bases, turning image matching into a problem of polynomial evaluation. The method is suited for coarse recognition between objects and, by adding more bases, also for fine recognition of the object pose. The advantages over the common representation using PCA include storing sampled points for recognition is not required, adding new objects to an existing data set is trivial because retraining other object models is not needed, and significantly in the important case where one has to scan an image over multiple locations in search for an object, the new representation is readily available as opposed to PCA projection at each location. The experimental result on the COIL-100 data set demonstrates high recognition accuracy with real-time performance.

Learning to detect objects in images via a sparse, part-based representation

We study the problem of detecting objects in still, gray-scale images. Our primary focus is the development of a learning-based approach to the problem that makes use of a sparse, part-based representation. A vocabulary of distinctive object parts is automatically constructed from a set of sample images of the object class of interest; images are then represented using parts from this vocabulary, together with spatial relations observed among the parts. Based on this representation, a learning algorithm is used to automatically learn to detect instances of the object class in new images. The approach can be applied to any object with distinguishable parts in a relatively fixed spatial configuration; it is evaluated here on difficult sets of real-world images containing side views of cars, and is seen to successfully detect objects in varying conditions amidst background clutter and mild occlusion. In evaluating object detection approaches, several important methodological issues arise that have not been satisfactorily addressed in previous work. A secondary focus of this paper is to highlight these issues and to develop rigorous evaluation standards for the object detection problem. A critical evaluation of our approach under the proposed standards is presented.

Pattern Filtering for Detection of Neural Activity, with Examples from HVc Activity During Sleep in Zebra Finches

The detection of patterned spiking activity is important in the study of neural coding. A pattern filtering approach is developed for pattern detection under the framework of point processes, which offers flexibility in combining temporal details and firing rates. The detection combines multiple steps of filtering in a coarse-to-fine manner. Under some conditional Poisson assumptions on the spiking activity, each filtering step is equivalent to classifying by likelihood ratios all the data segments as targets or as background sequences. Unlike previous studies, where global surrogate data were used to evaluate the statistical significance of the detected patterns, a localizedp-test procedure is developed, which better accounts for firing modulation and nonstationarity in spiking activity. Common temporal structures of patterned activity are learned using an entropy-based alignment procedure, without relying on metrics or pair-wise alignment. Applications of pattern filtering to single, presumptive interneurons recorded in the nucleus HVc of zebra finch are illustrated. These demonstrate a match between the auditory-evoked response to playback of the individual bird's own song and spontaneous activity during sleep. Small temporal compression or expansion, or both, is required for optimal matching of spontaneous patterns to stimulus-evoked activity.

Learning optimized features for hierarchical models of invariant object recognition

There is an ongoing debate over the capabilities of hierarchical neural feedforward architectures for performing real-world invariant object recognition. Although a variety of hierarchical models exists, appropriate supervised and unsupervised learning methods are still an issue of intense research. We propose a feedforward model for recognition that shares components like weight sharing, pooling stages, and competitive nonlinearities with earlier approaches but focuses on new methods for learning optimal feature-detecting cells in intermediate stages of the hierarchical network. We show that principles of sparse coding, which were previously mostly applied to the initial feature detection stages, can also be employed to obtain optimized intermediate complex features. We suggest a new approach to optimize the learning of sparse features under the constraints of a weight-sharing or convolutional architecture that uses pooling operations to achieve gradual invariance in the feature hierarchy. The approach explicitly enforces symmetry constraints like translation invariance on the feature set. This leads to a dimension reduction in the search space of optimal features and allows determining more efficiently the basis representatives, which achieve a sparse decomposition of the input. We analyze the quality of the learned feature representation by investigating the recognition performance of the resulting hierarchical network on object and face databases. We show that a hierarchy with features learned on a single object data set can also be applied to face recognition without parameter changes and is competitive with other recent machine learning recognition approaches. To investigate the effect of the interplay between sparse coding and processing nonlinearities, we also consider alternative feedforward pooling nonlinearities such as presynaptic maximum selection and sum-of-squares integration. The comparison shows that a combination of strong competitive nonlinearities with sparse coding offers the best recognition performance in the difficult scenario of segmentation-free recognition in cluttered surround. We demonstrate that for both learning and recognition, a precise segmentation of the objects is not necessary.