A feedforward architecture accounts for rapid categorization

  title={A feedforward architecture accounts for rapid categorization},
  author={Thomas Serre and Aude Oliva and Tomaso A. Poggio},
  journal={Proceedings of the National Academy of Sciences},
  pages={6424 - 6429}
Primates are remarkably good at recognizing objects. The level of performance of their visual system and its robustness to image degradations still surpasses the best computer vision systems despite decades of engineering effort. In particular, the high accuracy of primates in ultra rapid object categorization and rapid serial visual presentation tasks is remarkable. Given the number of processing stages involved and typical neural latencies, such rapid visual processing is likely to be mostly… 

Figures and Tables from this paper

Editorial: Hierarchical Object Representations in the Visual Cortex and Computer Vision
Over the past 40 years, Neurobiology and Computational Neuroscience have proved that deeper understanding of visual processes in humans and non-human primates can lead to important advancements in
How Deep is the Feature Analysis underlying Rapid Visual Categorization?
It is found that recognition accuracy increases with higher stages of visual processing but that human decisions agree best with predictions from intermediate stages, and that the complexity of visual representations afforded by modern deep network models may exceed those used by human participants during rapid categorization.
Vision: are models of object recognition catching up with the brain?
  • T. Poggio, S. Ullman
  • Computer Science, Biology
    Annals of the New York Academy of Sciences
  • 2013
The ongoing struggle of visual models to catch up with the visual cortex is discussed, key reasons for the relatively rapid improvement of artificial systems and models are identified, and open problems for computational vision in this domain are identified.
A quantitative theory of immediate visual recognition.
Invariance in Visual Object Recognition Requires Training: A Computational Argument
It is argued that this model is an interesting null-hypothesis to compare behavioral results with and concluded that it may explain several experimental findings.
Computational Models of Visual Object Recognition
Some of the initial steps toward a theoretical understanding of the computational principles behind transformation-invariant visual recognition in the primate cortex are summarized.
Invariant Recognition Shapes Neural Representations of Visual Input.
Results in brain imaging, neurophysiology, and computational neuroscience are reviewed in support of the hypothesis that the ability to support the invariant recognition of semantic entities in the visual world shapes which neural representations of sensory input are computed by human visual cortex.
Animal Detection Precedes Access to Scene Category
It is shown that animal – but not vehicle – detection clearly precedes scene categorization, and the idea that rapid animal detection might be based on early access of global scene statistics is challenged, and rather suggests a process based on the extraction of specific local complex features that might be hardwired in the visual system.
A rodent model for the study of invariant visual object recognition
The present work shows that rats possess more advanced visual abilities than previously appreciated and provides the first systematic evidence for invariant object recognition in rats, and argues for an increased focus on rodents as models for studying high-level visual processing.
Attentive processing improves object recognition
Using a Bayesian model of attention along with a hierarchical model of feed forward recognition on a data set of real world images, it is shown that this two stage attentive processing can improve recognition in cluttered and crowded conditions.


A Theory of Object Recognition: Computations and Circuits in the Feedforward Path of the Ventral Stream in Primate Visual Cortex
Abstract : We describe a quantitative theory to account for the computations performed by the feedforward path of the ventral stream of visual cortex and the local circuits implementing them. We show
Hierarchical models of object recognition in cortex
A new hierarchical model consistent with physiological data from inferotemporal cortex that accounts for this complex visual task and makes testable predictions is described.
Visual features of intermediate complexity and their use in classification
It is shown that intermediate complexity (IC) features are optimal for the basic visual task of classification and suggest a specific role for IC features in visual processing and a principle for their extraction.
Robust Object Recognition with Cortex-Like Mechanisms
A hierarchical system that closely follows the organization of visual cortex and builds an increasingly complex and invariant feature representation by alternating between a template matching and a maximum pooling operation is described.
Fast Readout of Object Identity from Macaque Inferior Temporal Cortex
Understanding the brain computations leading to object recognition requires quantitative characterization of the information represented in inferior temporal (IT) cortex. We used a biologically