Large Scale Fine-Grained Categorization and Domain-Specific Transfer Learning
Research output: Contribution to journal › Conference article › Research › peer-review
Transferring the knowledge learned from large scale datasets (e.g., ImageNet) via fine-tuning offers an effective solution for domain-specific fine-grained visual categorization (FGVC) tasks (e.g., recognizing bird species or car make & model). In such scenarios, data annotation often calls for specialized domain knowledge and thus is difficult to scale. In this work, we first tackle a problem in large scale FGVC. Our method won first place in iNaturalist 2017 large scale species classification challenge. Central to the success of our approach is a training scheme that uses higher image resolution and deals with the long-tailed distribution of training data. Next, we study transfer learning via fine-tuning from large scale datasets to small scale, domain-specific FGVC datasets. We propose a measure to estimate domain similarity via Earth Mover's Distance and demonstrate that transfer learning benefits from pre-training on a source domain that is similar to the target domain by this measure. Our proposed transfer learning outperforms ImageNet pre-training and obtains state-of-the-art results on multiple commonly used FGVC datasets.
Original language | English |
---|---|
Journal | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition |
Pages (from-to) | 4109-4118 |
Number of pages | 10 |
ISSN | 1063-6919 |
DOIs | |
Publication status | Published - 14 Dec 2018 |
Externally published | Yes |
Event | 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018 - Salt Lake City, United States Duration: 18 Jun 2018 → 22 Jun 2018 |
Conference
Conference | 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018 |
---|---|
Country | United States |
City | Salt Lake City |
Period | 18/06/2018 → 22/06/2018 |
Bibliographical note
Publisher Copyright:
© 2018 IEEE.
ID: 301825551