Deep Metric Learning via Facility Location
© 2017 IEEE. Learning image similarity metrics in an end-to-end fashion with deep networks has demonstrated excellent results on tasks such as clustering and retrieval. However, current methods, all focus on a very local view of the data. In this paper, we propose a new metric learning scheme, based...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021
|
| Online Access: | https://hdl.handle.net/1721.1/137678.2 |
| _version_ | 1826217709054984192 |
|---|---|
| author | Song, Hyun Oh Jegelka, Stefanie Sabrina Rathod, Vivek Murphy, Kevin |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Song, Hyun Oh Jegelka, Stefanie Sabrina Rathod, Vivek Murphy, Kevin |
| author_sort | Song, Hyun Oh |
| collection | MIT |
| description | © 2017 IEEE. Learning image similarity metrics in an end-to-end fashion with deep networks has demonstrated excellent results on tasks such as clustering and retrieval. However, current methods, all focus on a very local view of the data. In this paper, we propose a new metric learning scheme, based on structured prediction, that is aware of the global structure of the embedding space, and which is designed to optimize a clustering quality metric (NMI). We show state of the art performance on standard datasets, such as CUB200-2011 [37], Cars196 [18], and Stanford online products [30] on NMI and R@K evaluation metrics. |
| first_indexed | 2024-09-23T17:07:55Z |
| format | Article |
| id | mit-1721.1/137678.2 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T17:07:55Z |
| publishDate | 2021 |
| publisher | IEEE |
| record_format | dspace |
| spelling | mit-1721.1/137678.22021-12-20T20:12:46Z Deep Metric Learning via Facility Location Song, Hyun Oh Jegelka, Stefanie Sabrina Rathod, Vivek Murphy, Kevin Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory © 2017 IEEE. Learning image similarity metrics in an end-to-end fashion with deep networks has demonstrated excellent results on tasks such as clustering and retrieval. However, current methods, all focus on a very local view of the data. In this paper, we propose a new metric learning scheme, based on structured prediction, that is aware of the global structure of the embedding space, and which is designed to optimize a clustering quality metric (NMI). We show state of the art performance on standard datasets, such as CUB200-2011 [37], Cars196 [18], and Stanford online products [30] on NMI and R@K evaluation metrics. 2021-12-20T20:12:46Z 2021-11-08T15:12:27Z 2021-12-20T20:12:46Z 2017-07 2019-06-03T16:26:05Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137678.2 Song, Hyun Oh, Jegelka, Stefanie, Rathod, Vivek and Murphy, Kevin. 2017. "Deep Metric Learning via Facility Location." en 10.1109/cvpr.2017.237 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/octet-stream IEEE arXiv |
| spellingShingle | Song, Hyun Oh Jegelka, Stefanie Sabrina Rathod, Vivek Murphy, Kevin Deep Metric Learning via Facility Location |
| title | Deep Metric Learning via Facility Location |
| title_full | Deep Metric Learning via Facility Location |
| title_fullStr | Deep Metric Learning via Facility Location |
| title_full_unstemmed | Deep Metric Learning via Facility Location |
| title_short | Deep Metric Learning via Facility Location |
| title_sort | deep metric learning via facility location |
| url | https://hdl.handle.net/1721.1/137678.2 |
| work_keys_str_mv | AT songhyunoh deepmetriclearningviafacilitylocation AT jegelkastefaniesabrina deepmetriclearningviafacilitylocation AT rathodvivek deepmetriclearningviafacilitylocation AT murphykevin deepmetriclearningviafacilitylocation |