Subtasks of unconstrained face recognition
Unconstrained face recognition remains a challenging computer vision problem despite recent exceptionally high results ( ~ 95% accuracy) on the current gold standard evaluation dataset: Labeled Faces in the Wild (LFW). We offer a decomposition of the unconstrained problem into subtasks based on the...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
2016
|
| Online Access: | http://hdl.handle.net/1721.1/102486 https://orcid.org/0000-0002-3153-916X https://orcid.org/0000-0002-3944-0455 https://orcid.org/0000-0003-0076-621X |
| _version_ | 1826201052032008192 |
|---|---|
| author | Leibo, Joel Z. Liao, Qianli Poggio, Tomaso A. |
| author2 | Center for Brains, Minds and Machines at MIT |
| author_facet | Center for Brains, Minds and Machines at MIT Leibo, Joel Z. Liao, Qianli Poggio, Tomaso A. |
| author_sort | Leibo, Joel Z. |
| collection | MIT |
| description | Unconstrained face recognition remains a challenging computer vision problem despite recent exceptionally high results ( ~ 95% accuracy) on the current gold standard evaluation dataset: Labeled Faces in the Wild (LFW). We offer a decomposition of the unconstrained problem into subtasks based on the idea that invariance to identity-preserving transformations is the crux of recognition. Each of the subtasks in the Subtasks of Unconstrained Face Recognition (SUFR) challenge consists of a same-different face-matching problem on a set of 400 individual synthetic faces rendered so as to isolate a specific transformation or set of transformations. We characterized the performance of 9 different models (8 previously published) on each of the subtasks. One notable finding was that the HMAX-C2 feature was not nearly as clutter-resistant as had been suggested by previous publications. Next we considered LFW and argued that it is too easy of a task to continue to be regarded as a measure of progress on unconstrained face recognition. In particular, strong performance on LFW requires almost no invariance, yet it cannot be considered a fair approximation of the outcome of a detection --> alignment pipeline since it does not contain the kinds of variability that realistic alignment systems produce when working on non-frontal faces. We offer a new, more difficult, natural image dataset: SUFR-in-the-Wild (SUFR-W), which we created using a protocol that was similar to LFW, but with a few differences designed to produce more need for transformation invariance. We present baseline results for eight different face recognition systems on the new dataset and argue that it is time to retire LFW and move on to more difficult evaluations for unconstrained face recognition. |
| first_indexed | 2024-09-23T11:45:54Z |
| format | Article |
| id | mit-1721.1/102486 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T11:45:54Z |
| publishDate | 2016 |
| record_format | dspace |
| spelling | mit-1721.1/1024862022-10-01T05:52:23Z Subtasks of unconstrained face recognition Leibo, Joel Z. Liao, Qianli Poggio, Tomaso A. Center for Brains, Minds and Machines at MIT Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science McGovern Institute for Brain Research at MIT Leibo, Joel Z. Liao, Qianli Poggio, Tomaso A. Unconstrained face recognition remains a challenging computer vision problem despite recent exceptionally high results ( ~ 95% accuracy) on the current gold standard evaluation dataset: Labeled Faces in the Wild (LFW). We offer a decomposition of the unconstrained problem into subtasks based on the idea that invariance to identity-preserving transformations is the crux of recognition. Each of the subtasks in the Subtasks of Unconstrained Face Recognition (SUFR) challenge consists of a same-different face-matching problem on a set of 400 individual synthetic faces rendered so as to isolate a specific transformation or set of transformations. We characterized the performance of 9 different models (8 previously published) on each of the subtasks. One notable finding was that the HMAX-C2 feature was not nearly as clutter-resistant as had been suggested by previous publications. Next we considered LFW and argued that it is too easy of a task to continue to be regarded as a measure of progress on unconstrained face recognition. In particular, strong performance on LFW requires almost no invariance, yet it cannot be considered a fair approximation of the outcome of a detection --> alignment pipeline since it does not contain the kinds of variability that realistic alignment systems produce when working on non-frontal faces. We offer a new, more difficult, natural image dataset: SUFR-in-the-Wild (SUFR-W), which we created using a protocol that was similar to LFW, but with a few differences designed to produce more need for transformation invariance. We present baseline results for eight different face recognition systems on the new dataset and argue that it is time to retire LFW and move on to more difficult evaluations for unconstrained face recognition. 2016-05-13T19:00:14Z 2016-05-13T19:00:14Z 2014-01 Article http://purl.org/eprint/type/ConferencePaper http://hdl.handle.net/1721.1/102486 Leibo, Joel Z., Qianli Liao, and Tomaso Poggio. "Subtasks of unconstrained face recognition." 2014 9th International Conference on Computer Vision Theory and Applications (VISAPP 2014) (January 2014). https://orcid.org/0000-0002-3153-916X https://orcid.org/0000-0002-3944-0455 https://orcid.org/0000-0003-0076-621X en_US http://www.visapp.visigrapp.org/Abstracts/2014/VISAPP_2014_Abstracts.htm Proceedings of the 2014 9th International Conference on Computer Vision Theory and Applications (VISAPP 2014) Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf MIT web domain |
| spellingShingle | Leibo, Joel Z. Liao, Qianli Poggio, Tomaso A. Subtasks of unconstrained face recognition |
| title | Subtasks of unconstrained face recognition |
| title_full | Subtasks of unconstrained face recognition |
| title_fullStr | Subtasks of unconstrained face recognition |
| title_full_unstemmed | Subtasks of unconstrained face recognition |
| title_short | Subtasks of unconstrained face recognition |
| title_sort | subtasks of unconstrained face recognition |
| url | http://hdl.handle.net/1721.1/102486 https://orcid.org/0000-0002-3153-916X https://orcid.org/0000-0002-3944-0455 https://orcid.org/0000-0003-0076-621X |
| work_keys_str_mv | AT leibojoelz subtasksofunconstrainedfacerecognition AT liaoqianli subtasksofunconstrainedfacerecognition AT poggiotomasoa subtasksofunconstrainedfacerecognition |