OTA: Optimal Transport Assignment for Object Detection

Zheng Ge, Songtao Liu*, Zeming Li, Osamu Yoshie, Jian Sun

*この研究の対応する著者

研究成果: Conference contribution

10 被引用数 (Scopus)

抄録

Recent advances in label assignment in object detection mainly seek to independently define positive/negative training samples for each ground-truth (gt) object. In this paper, we innovatively revisit the label assignment from a global perspective and propose to formulate the assigning procedure as an Optimal Transport (OT) problem - a well-studied topic in Optimization Theory. Concretely, we define the unit transportation cost between each demander (anchor) and supplier (gt) pair as the weighted summation of their classification and regression losses. After formulation, finding the best assignment solution is converted to solve the optimal transport plan at minimal transportation costs, which can be solved via Sinkhorn-Knopp Iteration. On COCO, a single FCOS-ResNet-50 detector equipped with Optimal Transport Assignment (OTA) can reach 40.7% mAP under 1× scheduler, outperforming all other existing assigning methods. Extensive experiments conducted on COCO and CrowdHuman further validate the effectiveness of our proposed OTA, especially its superiority in crowd scenarios. The code is available at https://github.com/Megvii-BaseDetection/OTA.

本文言語English
ホスト出版物のタイトルProceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
出版社IEEE Computer Society
ページ303-312
ページ数10
ISBN(電子版)9781665445092
DOI
出版ステータスPublished - 2021
イベント2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021 - Virtual, Online, United States
継続期間: 2021 6月 192021 6月 25

出版物シリーズ

名前Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN(印刷版)1063-6919

Conference

Conference2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
国/地域United States
CityVirtual, Online
Period21/6/1921/6/25

ASJC Scopus subject areas

  • ソフトウェア
  • コンピュータ ビジョンおよびパターン認識

フィンガープリント

「OTA: Optimal Transport Assignment for Object Detection」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル