Eccentricity dependent deep neural networks: Modeling invariance in human vision
Humans can recognize objects in a way that is invariant to scale, translation, and clutter. We use invariance theory as a conceptual basis, to computationally model this phenomenon. This theory discusses the role of eccentricity in human visual processing, and is a generalization of feedforward conv...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Association for the Advancement of Artificial Intelligence
2017
|
| Online Access: | http://hdl.handle.net/1721.1/112279 https://orcid.org/0000-0002-1909-257X https://orcid.org/0000-0002-7470-0179 https://orcid.org/0000-0002-9255-0151 https://orcid.org/0000-0002-3944-0455 |
| _version_ | 1811080221880745984 |
|---|---|
| author | Chen, Francis X. Roig Noguera, Gemma Isik, Leyla Boix Bosch, Xavier Poggio, Tomaso A |
| author2 | Center for Brains, Minds, and Machines |
| author_facet | Center for Brains, Minds, and Machines Chen, Francis X. Roig Noguera, Gemma Isik, Leyla Boix Bosch, Xavier Poggio, Tomaso A |
| author_sort | Chen, Francis X. |
| collection | MIT |
| description | Humans can recognize objects in a way that is invariant to scale, translation, and clutter. We use invariance theory as a conceptual basis, to computationally model this phenomenon. This theory discusses the role of eccentricity in human visual processing, and is a generalization of feedforward convolutional neural networks (CNNs). Our model explains some key psychophysical observations relating to invariant perception, while maintaining important similarities with biological neural architectures. To our knowledge, this work is the first to unify explanations of all three types of invariance, all while leveraging the power and neurological grounding of CNNs. |
| first_indexed | 2024-09-23T11:27:46Z |
| format | Article |
| id | mit-1721.1/112279 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T11:27:46Z |
| publishDate | 2017 |
| publisher | Association for the Advancement of Artificial Intelligence |
| record_format | dspace |
| spelling | mit-1721.1/1122792022-10-01T03:47:42Z Eccentricity dependent deep neural networks: Modeling invariance in human vision Chen, Francis X. Roig Noguera, Gemma Isik, Leyla Boix Bosch, Xavier Poggio, Tomaso A Center for Brains, Minds, and Machines Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Chen, Francis X. Roig Noguera, Gemma Isik, Leyla Boix Bosch, Xavier Poggio, Tomaso A Humans can recognize objects in a way that is invariant to scale, translation, and clutter. We use invariance theory as a conceptual basis, to computationally model this phenomenon. This theory discusses the role of eccentricity in human visual processing, and is a generalization of feedforward convolutional neural networks (CNNs). Our model explains some key psychophysical observations relating to invariant perception, while maintaining important similarities with biological neural architectures. To our knowledge, this work is the first to unify explanations of all three types of invariance, all while leveraging the power and neurological grounding of CNNs. 2017-11-22T16:03:27Z 2017-11-22T16:03:27Z 2017-03 2017-11-16T19:35:26Z Article http://purl.org/eprint/type/ConferencePaper http://hdl.handle.net/1721.1/112279 Chen,Francis X. et al. "Eccentricity dependent deep neural networks: Modeling invariance in human vision." 2017 AAAI Spring Symposium Series, Science of Intelligence: Computational Principles of Natural and Artificial Intelligence, March 27-29 2017, Stanford, California, Association for the Advancement of Artificial Intelligence, March 2017 © 2017 Association for the Advancement of Artificial Intelligence https://orcid.org/0000-0002-1909-257X https://orcid.org/0000-0002-7470-0179 https://orcid.org/0000-0002-9255-0151 https://orcid.org/0000-0002-3944-0455 https://www.aaai.org/ocs/index.php/SSS/SSS17/paper/viewPaper/15360 2017 AAAI Spring Symposium Series, Science of Intelligence: Computational Principles of Natural and Artificial Intelligence Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Association for the Advancement of Artificial Intelligence MIT Web Domain |
| spellingShingle | Chen, Francis X. Roig Noguera, Gemma Isik, Leyla Boix Bosch, Xavier Poggio, Tomaso A Eccentricity dependent deep neural networks: Modeling invariance in human vision |
| title | Eccentricity dependent deep neural networks: Modeling invariance in human vision |
| title_full | Eccentricity dependent deep neural networks: Modeling invariance in human vision |
| title_fullStr | Eccentricity dependent deep neural networks: Modeling invariance in human vision |
| title_full_unstemmed | Eccentricity dependent deep neural networks: Modeling invariance in human vision |
| title_short | Eccentricity dependent deep neural networks: Modeling invariance in human vision |
| title_sort | eccentricity dependent deep neural networks modeling invariance in human vision |
| url | http://hdl.handle.net/1721.1/112279 https://orcid.org/0000-0002-1909-257X https://orcid.org/0000-0002-7470-0179 https://orcid.org/0000-0002-9255-0151 https://orcid.org/0000-0002-3944-0455 |
| work_keys_str_mv | AT chenfrancisx eccentricitydependentdeepneuralnetworksmodelinginvarianceinhumanvision AT roignogueragemma eccentricitydependentdeepneuralnetworksmodelinginvarianceinhumanvision AT isikleyla eccentricitydependentdeepneuralnetworksmodelinginvarianceinhumanvision AT boixboschxavier eccentricitydependentdeepneuralnetworksmodelinginvarianceinhumanvision AT poggiotomasoa eccentricitydependentdeepneuralnetworksmodelinginvarianceinhumanvision |