Efficient feedforward categorization of objects and human postures with address-event image sensors
This paper proposes an algorithm for feedforward categorization of objects and, in particular, human postures in real-time video sequences from address-event temporal-difference image sensors. The system employs an innovative combination of event-based hardware and bio-inspired software architecture...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/99187 http://hdl.handle.net/10220/13517 |
| _version_ | 1824455650163294208 |
|---|---|
| author | Chen, Shoushun Akselrod, Polina Zhao, Bo Carrasco, Jose Antonio Perez Linares-Barranco, Bernabe Culurciello, Eugenio |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Chen, Shoushun Akselrod, Polina Zhao, Bo Carrasco, Jose Antonio Perez Linares-Barranco, Bernabe Culurciello, Eugenio |
| author_sort | Chen, Shoushun |
| collection | NTU |
| description | This paper proposes an algorithm for feedforward categorization of objects and, in particular, human postures in real-time video sequences from address-event temporal-difference image sensors. The system employs an innovative combination of event-based hardware and bio-inspired software architecture. An event-based temporal difference image sensor is used to provide input video sequences, while a software module extracts size and position invariant line features inspired by models of the primate visual cortex. The detected line features are organized into vectorial segments. After feature extraction, a modified line segment Hausdorff-distance classifier combined with on-the-fly cluster-based size and position invariant categorization. The system can achieve about 90 percent average success rate in the categorization of human postures, while using only a small number of training samples. Compared to state-of-the-art bio-inspired categorization methods, the proposed algorithm requires less hardware resource, reduces the computation complexity by at least five times, and is an ideal candidate for hardware implementation with event-based circuits. |
| first_indexed | 2025-02-19T03:41:34Z |
| format | Journal Article |
| id | ntu-10356/99187 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:41:34Z |
| publishDate | 2013 |
| record_format | dspace |
| spelling | ntu-10356/991872020-03-07T13:57:23Z Efficient feedforward categorization of objects and human postures with address-event image sensors Chen, Shoushun Akselrod, Polina Zhao, Bo Carrasco, Jose Antonio Perez Linares-Barranco, Bernabe Culurciello, Eugenio School of Electrical and Electronic Engineering DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision This paper proposes an algorithm for feedforward categorization of objects and, in particular, human postures in real-time video sequences from address-event temporal-difference image sensors. The system employs an innovative combination of event-based hardware and bio-inspired software architecture. An event-based temporal difference image sensor is used to provide input video sequences, while a software module extracts size and position invariant line features inspired by models of the primate visual cortex. The detected line features are organized into vectorial segments. After feature extraction, a modified line segment Hausdorff-distance classifier combined with on-the-fly cluster-based size and position invariant categorization. The system can achieve about 90 percent average success rate in the categorization of human postures, while using only a small number of training samples. Compared to state-of-the-art bio-inspired categorization methods, the proposed algorithm requires less hardware resource, reduces the computation complexity by at least five times, and is an ideal candidate for hardware implementation with event-based circuits. 2013-09-18T03:54:08Z 2019-12-06T20:04:15Z 2013-09-18T03:54:08Z 2019-12-06T20:04:15Z 2012 2012 Journal Article Chen, S., Akseirod, P., Zhao, B., Carrasco, J. A. P., Linares-Barranco, B., & Culurciello, E. (2012). Efficient feedforward categorization of objects and human postures with address-event image sensors. IEEE transactions on pattern analysis and machine intelligence, 34(2), 302-314. 0162-8828 https://hdl.handle.net/10356/99187 http://hdl.handle.net/10220/13517 10.1109/TPAMI.2011.120 en IEEE transactions on pattern analysis and machine intelligence © 2012 IEEE |
| spellingShingle | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Chen, Shoushun Akselrod, Polina Zhao, Bo Carrasco, Jose Antonio Perez Linares-Barranco, Bernabe Culurciello, Eugenio Efficient feedforward categorization of objects and human postures with address-event image sensors |
| title | Efficient feedforward categorization of objects and human postures with address-event image sensors |
| title_full | Efficient feedforward categorization of objects and human postures with address-event image sensors |
| title_fullStr | Efficient feedforward categorization of objects and human postures with address-event image sensors |
| title_full_unstemmed | Efficient feedforward categorization of objects and human postures with address-event image sensors |
| title_short | Efficient feedforward categorization of objects and human postures with address-event image sensors |
| title_sort | efficient feedforward categorization of objects and human postures with address event image sensors |
| topic | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision |
| url | https://hdl.handle.net/10356/99187 http://hdl.handle.net/10220/13517 |
| work_keys_str_mv | AT chenshoushun efficientfeedforwardcategorizationofobjectsandhumanpostureswithaddresseventimagesensors AT akselrodpolina efficientfeedforwardcategorizationofobjectsandhumanpostureswithaddresseventimagesensors AT zhaobo efficientfeedforwardcategorizationofobjectsandhumanpostureswithaddresseventimagesensors AT carrascojoseantonioperez efficientfeedforwardcategorizationofobjectsandhumanpostureswithaddresseventimagesensors AT linaresbarrancobernabe efficientfeedforwardcategorizationofobjectsandhumanpostureswithaddresseventimagesensors AT culurcielloeugenio efficientfeedforwardcategorizationofobjectsandhumanpostureswithaddresseventimagesensors |