CaT: Weakly Supervised Object Detection With Category Transfer

Tianyue Cao, Lianyu Du, Xiaoyun Zhang, Siheng Chen, Ya Zhang, Yan-Feng Wang; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021, pp. 3070-3079

Abstract


A large gap exists between fully-supervised object detection and weakly-supervised object detection. To narrow this gap, some methods consider knowledge transfer from additional fully-supervised dataset. But these methods do not fully exploit discriminative category information in the fully-supervised dataset, thus causing low mAP. To solve this issue, we propose a novel category transfer framework for weakly supervised object detection. The intuition is to fully leverage both visually-discriminative and semantically-correlated category information in the fully-supervised dataset to enhance the object-classification ability of a weakly-supervised detector. To handle overlapping category transfer, we propose a double-supervision mean teacher to gather common category information and bridge the domain gap between two datasets. To handle non-overlapping category transfer, we propose a semantic graph convolutional network to promote the aggregation of semantic features between correlated categories. Experiments are conducted with Pascal VOC 2007 as the target weakly-supervised dataset and COCO as the source fully-supervised dataset. Our category transfer framework achieves 63.5% mAP and 80.3% CorLoc with 5 overlapping categories between two datasets, which outperforms the state-of-the-art methods. Codes are avaliable at https://github.com/MediaBrain-SJTU/CaT.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Cao_2021_ICCV, author = {Cao, Tianyue and Du, Lianyu and Zhang, Xiaoyun and Chen, Siheng and Zhang, Ya and Wang, Yan-Feng}, title = {CaT: Weakly Supervised Object Detection With Category Transfer}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}, month = {October}, year = {2021}, pages = {3070-3079} }