Approximations in the HMAX Model
The HMAX model is a biologically motivated architecture for computer vision whose components are in close agreement with existing physiological evidence. The model is capable of achieving close to human level performance on several rapid object recognition tasks. However, the model is computationall...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Published: |
2011
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/62293 |
| _version_ | 1826204508552691712 |
|---|---|
| author | Chikkerur, Sharat Poggio, Tomaso |
| author2 | Tomaso Poggio |
| author_facet | Tomaso Poggio Chikkerur, Sharat Poggio, Tomaso |
| author_sort | Chikkerur, Sharat |
| collection | MIT |
| description | The HMAX model is a biologically motivated architecture for computer vision whose components are in close agreement with existing physiological evidence. The model is capable of achieving close to human level performance on several rapid object recognition tasks. However, the model is computationally bound and has limited engineering applications in its current form. In this report, we present several approximations in order to increase the efficiency of the HMAX model. We outline approximations at several levels of the hierarchy and empirically evaluate the trade-offs between efficiency and accuracy. We also explore ways to quantify the representation capacity of the model. |
| first_indexed | 2024-09-23T12:56:43Z |
| id | mit-1721.1/62293 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T12:56:43Z |
| publishDate | 2011 |
| record_format | dspace |
| spelling | mit-1721.1/622932019-04-12T12:45:33Z Approximations in the HMAX Model Chikkerur, Sharat Poggio, Tomaso Tomaso Poggio Center for Biological and Computational Learning (CBCL) object recognition, approximation The HMAX model is a biologically motivated architecture for computer vision whose components are in close agreement with existing physiological evidence. The model is capable of achieving close to human level performance on several rapid object recognition tasks. However, the model is computationally bound and has limited engineering applications in its current form. In this report, we present several approximations in order to increase the efficiency of the HMAX model. We outline approximations at several levels of the hierarchy and empirically evaluate the trade-offs between efficiency and accuracy. We also explore ways to quantify the representation capacity of the model. 2011-04-21T18:15:06Z 2011-04-21T18:15:06Z 2011-04-14 http://hdl.handle.net/1721.1/62293 MIT-CSAIL-TR-2011-021 CBCL-298 Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported http://creativecommons.org/licenses/by-nc-nd/3.0/ 12 p. application/pdf |
| spellingShingle | object recognition, approximation Chikkerur, Sharat Poggio, Tomaso Approximations in the HMAX Model |
| title | Approximations in the HMAX Model |
| title_full | Approximations in the HMAX Model |
| title_fullStr | Approximations in the HMAX Model |
| title_full_unstemmed | Approximations in the HMAX Model |
| title_short | Approximations in the HMAX Model |
| title_sort | approximations in the hmax model |
| topic | object recognition, approximation |
| url | http://hdl.handle.net/1721.1/62293 |
| work_keys_str_mv | AT chikkerursharat approximationsinthehmaxmodel AT poggiotomaso approximationsinthehmaxmodel |