On Label Granularity and Object Localization
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Weakly supervised object localization (WSOL) aims to learn representations that encode object location using only image-level category labels. However, many objects can be labeled at different levels of granularity. Is it an animal, a bird, or a great horned owl? Which image-level labels should we use? In this paper we study the role of label granularity in WSOL. To facilitate this investigation we introduce iNatLoc500, a new large-scale fine-grained benchmark dataset for WSOL. Surprisingly, we find that choosing the right training label granularity provides a much larger performance boost than choosing the best WSOL algorithm. We also show that changing the label granularity can significantly improve data efficiency.
Originalsprog | Engelsk |
---|---|
Titel | Computer Vision – ECCV 2022 : 17th European Conference, Proceedings |
Redaktører | Shai Avidan, Gabriel Brostow, Moustapha Cissé, Giovanni Maria Farinella, Tal Hassner |
Antal sider | 17 |
Forlag | Springer |
Publikationsdato | 2022 |
Sider | 604-620 |
ISBN (Trykt) | 9783031200793 |
DOI | |
Status | Udgivet - 2022 |
Begivenhed | 17th European Conference on Computer Vision, ECCV 2022 - Tel Aviv, Israel Varighed: 23 okt. 2022 → 27 okt. 2022 |
Konference
Konference | 17th European Conference on Computer Vision, ECCV 2022 |
---|---|
Land | Israel |
By | Tel Aviv |
Periode | 23/10/2022 → 27/10/2022 |
Navn | Lecture Notes in Computer Science |
---|---|
Vol/bind | 13670 LNCS |
ISSN | 0302-9743 |
Bibliografisk note
Funding Information:
Acknowledgements. We thank the iNaturalist community for sharing images and species annotations. This work was supported by the Caltech Resnick Sustainability Institute, an NSF Graduate Research Fellowship (grant number DGE1745301), and the Pioneer Centre for AI (DNRF grant number P1).
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Links
- https://arxiv.org/pdf/2207.10225
Indsendt manuskript
ID: 342672283