A Computational Model for Visual Selection

@article{Amit1999ACM,
  title={A Computational Model for Visual Selection},
  author={Yali Amit and Donald Geman},
  journal={Neural Computation},
  year={1999},
  volume={11},
  pages={1691-1715}
}
  • Y. Amit, D. Geman
  • Published 1 October 1999
  • Computer Science
  • Neural Computation
We propose a computational model for detecting and localizing instances from an object class in static gray-level images. We divide detection into visual selection and final classification, concentrating on the former: drastically reducing the number of candidate regions that require further, usually more intensive, processing, but with a minimum of computation and missed detections. Bottom-up processing is based on local groupings of edge fragments constrained by loose geometrical… 
Coarse-to-Fine Visual Selection
TLDR
The approach is sequential testing which is coarse-tone in both in the exploration of poses and the representation of objects, and the spatial distribution of processing is highly skewed and detection is rapid, but at the expense of false alarms which, presumably, could be eliminated with localized, more intensive, processing.
Coarse-to-Fine Face Detection
TLDR
The approach is sequential testing which is coarse-to-fine in both in the exploration of poses and the representation of objects, and the spatial distribution of processing is highly skewed and detection is rapid, but at the expense of (isolated) false alarms which could be eliminated with localized, more intensive, processing.
A Neural Network Architecture for Visual Selection
  • Y. Amit
  • Computer Science
    Neural Computation
  • 2000
TLDR
A parallel neural net architecture for efficient and robust visual selection in generic gray-level images, constructed so that the Hough transform associated with any object can be implemented without creating or modifying any connections.
Computational Strategies for Model-Based Scene Interpretation
Scene interpretation, in the sense of detecting and localizing instances from multiple object classes, is formulated as a two-step process in which non-contextual detection primes global
Coarse-to-Fine Classification and Scene Labeling
TLDR
The models and algorithms presented here result from making computational efficiency the organizing principle for vision, a proposal recently explored in both theory and practice, and some results about optimal testing strategies are summarized.
2D Object Detection and Recognition: Models, Algorithms, and Networks
TLDR
This book discusses the construction and training of models, computational approaches to efficient implementation, and parallel implementations in biologically plausible neural network architectures based on statistical modeling and estimation, with an emphasis on simplicity, transparency, and computational efficiency.
Acquisition of visual shape primitives
An integrated network for invariant visual detection and recognition
On the Optimality of Spatial Attention for Object Detection
TLDR
A simple, formal mathematical model of the advantage of spatial attention for object detection is developed, in which spatial attention is defined as processing a subset of the visual input, and detection is an abstraction with certain failure characteristics.
Graded Learning for Object Detection
TLDR
The goal is to detect all instances of a generic object class, such as a face, in greyscale scenes by learning a hierarchy of spatial arrangements of edge fragments, graded by their size (sparsity).
...
...

References

SHOWING 1-10 OF 46 REFERENCES
A Neural Network Architecture for Visual Selection
  • Y. Amit
  • Computer Science
    Neural Computation
  • 2000
TLDR
A parallel neural net architecture for efficient and robust visual selection in generic gray-level images, constructed so that the Hough transform associated with any object can be implemented without creating or modifying any connections.
Human image understanding: Recent research and a theory
  • I. Biederman
  • Computer Science
    Comput. Vis. Graph. Image Process.
  • 1985
Inferior temporal mechanisms for invariant object recognition.
TLDR
IT neuronal populations may mediate not only the recognition and memory of object identity, which are invariant over size and location, but also the perception of the transformations themselves.
Shape Quantization and Recognition with Randomized Trees
TLDR
A new approach to shape recognition based on a virtually infinite family of binary features (queries) of the image data, designed to accommodate prior information about shape invariance and regularity, and a comparison with artificial neural networks methods is presented.
Macaque VI neurons can signal ‘illusory’ contours
TLDR
There are neurons as early as V1 with the computational power to detect illusory contour and to help distinguish figure from ground, according to a new view of primary visual cortex based on measurements of neural responses in V1 to patterns called ‘illusory contours’.
Psychophysical support for a 2D view interpolation theory of object recognition
TLDR
The results suggest that the human visual system is better described as recognizing these objects by 2D view interpolation than by alignment or other methods that rely on object-centered 3D models, as well as those recognition theories that represent objects by 3D structural relationships between generic volumetric primitives.
Psychophysical support for a two-dimensional view interpolation theory of object recognition.
  • H. Bülthoff, S. Edelman
  • Computer Science
    Proceedings of the National Academy of Sciences of the United States of America
  • 1992
TLDR
The results suggest that the human visual system is better described as recognizing these objects by two-dimensional view interpolation than by alignment or other methods that rely on object-centered three-dimensional models.
Columns for visual features of objects in monkey inferotemporal cortex
TLDR
The anterior inferotemporal area of the monkey cortex, the final station of the visual cortical stream crucial for object recognition, consists of columns, each containing cells responsive to similar visual features of objects.
Efficient Focusing and Face Detection
TLDR
An algorithm for shape detection is presented and applied to frontal views of faces in still grey level images with arbitrary backgrounds and a selected region-of-interest is labeled face or background based on multiple decision trees and normalized data.
Neuronal selectivities to complex object features in the ventral visual pathway of the macaque cerebral cortex.
TLDR
It is suggested that local neuronal networks in V4 and posterior IT play an essential role in the formation of selective responses to complex object features.
...
...