A new biologically motivated framework for robust object recognition
In this paper, we introduce a novel set of features for robust object recognition, which exhibits outstanding performances on a variety ofobject categories while being capable of learning from only a fewtraining examples. Each element of this set is a complex featureobtained by combining position- a...
| Main Authors: | , , |
|---|---|
| Language: | en_US |
| Published: |
2005
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/30504 |
| _version_ | 1826217981450911744 |
|---|---|
| author | Serre, Thomas Wolf, Lior Poggio, Tomaso |
| author_facet | Serre, Thomas Wolf, Lior Poggio, Tomaso |
| author_sort | Serre, Thomas |
| collection | MIT |
| description | In this paper, we introduce a novel set of features for robust object recognition, which exhibits outstanding performances on a variety ofobject categories while being capable of learning from only a fewtraining examples. Each element of this set is a complex featureobtained by combining position- and scale-tolerant edge-detectors overneighboring positions and multiple orientations.Our system - motivated by a quantitative model of visual cortex -outperforms state-of-the-art systems on a variety of object imagedatasets from different groups. We also show that our system is ableto learn from very few examples with no prior category knowledge. Thesuccess of the approach is also a suggestive plausibility proof for aclass of feed-forward models of object recognition in cortex. Finally,we conjecture the existence of a universal overcompletedictionary of features that could handle the recognition of all objectcategories. |
| first_indexed | 2024-09-23T17:12:16Z |
| id | mit-1721.1/30504 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T17:12:16Z |
| publishDate | 2005 |
| record_format | dspace |
| spelling | mit-1721.1/305042019-04-12T08:37:51Z A new biologically motivated framework for robust object recognition Serre, Thomas Wolf, Lior Poggio, Tomaso AI visual cortex object recognition face detection hierarchy feature learning In this paper, we introduce a novel set of features for robust object recognition, which exhibits outstanding performances on a variety ofobject categories while being capable of learning from only a fewtraining examples. Each element of this set is a complex featureobtained by combining position- and scale-tolerant edge-detectors overneighboring positions and multiple orientations.Our system - motivated by a quantitative model of visual cortex -outperforms state-of-the-art systems on a variety of object imagedatasets from different groups. We also show that our system is ableto learn from very few examples with no prior category knowledge. Thesuccess of the approach is also a suggestive plausibility proof for aclass of feed-forward models of object recognition in cortex. Finally,we conjecture the existence of a universal overcompletedictionary of features that could handle the recognition of all objectcategories. 2005-12-22T02:15:45Z 2005-12-22T02:15:45Z 2004-11-14 MIT-CSAIL-TR-2004-074 AIM-2004-026 CBCL-243 http://hdl.handle.net/1721.1/30504 en_US Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory 10 p. 17638397 bytes 793841 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | AI visual cortex object recognition face detection hierarchy feature learning Serre, Thomas Wolf, Lior Poggio, Tomaso A new biologically motivated framework for robust object recognition |
| title | A new biologically motivated framework for robust object recognition |
| title_full | A new biologically motivated framework for robust object recognition |
| title_fullStr | A new biologically motivated framework for robust object recognition |
| title_full_unstemmed | A new biologically motivated framework for robust object recognition |
| title_short | A new biologically motivated framework for robust object recognition |
| title_sort | new biologically motivated framework for robust object recognition |
| topic | AI visual cortex object recognition face detection hierarchy feature learning |
| url | http://hdl.handle.net/1721.1/30504 |
| work_keys_str_mv | AT serrethomas anewbiologicallymotivatedframeworkforrobustobjectrecognition AT wolflior anewbiologicallymotivatedframeworkforrobustobjectrecognition AT poggiotomaso anewbiologicallymotivatedframeworkforrobustobjectrecognition AT serrethomas newbiologicallymotivatedframeworkforrobustobjectrecognition AT wolflior newbiologicallymotivatedframeworkforrobustobjectrecognition AT poggiotomaso newbiologicallymotivatedframeworkforrobustobjectrecognition |