Recurrent Connections Aid Occluded Object Recognition by Discounting Occluders
| Autor: | Thomas Burwick, Jochen Triesch, Markus Roland Ernst |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Computer science Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Stereoscopy Stimulus (physiology) 050105 experimental psychology law.invention Machine Learning (cs.LG) 03 medical and health sciences 0302 clinical medicine law medicine FOS: Electrical engineering electronic engineering information engineering 0501 psychology and cognitive sciences Parametric statistics Artificial neural network business.industry 05 social sciences Image and Video Processing (eess.IV) Cognitive neuroscience of visual object recognition Feed forward Pattern recognition Electrical Engineering and Systems Science - Image and Video Processing Visual cortex medicine.anatomical_structure Artificial intelligence business 030217 neurology & neurosurgery |
| Zdroj: | Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing-28th International Conference on Artificial Neural Networks, Munich, Germany, September 17–19, 2019, Proceedings, Part III Lecture Notes in Computer Science Lecture Notes in Computer Science-Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing Lecture Notes in Computer Science ISBN: 9783030305079 |
| ISSN: | 0302-9743 1611-3349 |
| DOI: | 10.48550/arxiv.1907.08831 |
| Popis: | Recurrent connections in the visual cortex are thought to aid object recognition when part of the stimulus is occluded. Here we investigate if and how recurrent connections in artificial neural networks similarly aid object recognition. We systematically test and compare architectures comprised of bottom-up (B), lateral (L) and top-down (T) connections. Performance is evaluated on a novel stereoscopic occluded object recognition dataset. The task consists of recognizing one target digit occluded by multiple occluder digits in a pseudo-3D environment. We find that recurrent models perform significantly better than their feedforward counterparts, which were matched in parametric complexity. Furthermore, we analyze how the network's representation of the stimuli evolves over time due to recurrent connections. We show that the recurrent connections tend to move the network's representation of an occluded digit towards its un-occluded version. Our results suggest that both the brain and artificial neural networks can exploit recurrent connectivity to aid occluded object recognition. Comment: 13 pages, 5 figures, accepted at the 28th International Conference on Artificial Neural Networks, published in Springer Lecture Notes in Computer Science vol 11729 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|