UWSOD: Toward Fully-Supervised-Level Capacity Weakly Supervised Object Detection
NIPS 2020, 2020.
We propose a unified weakly supervised object detection framework, termed UWSOD, to develop a high-capacity general detection model with only image-level labels, which is self-contained and does not require external modules or additional supervision
Weakly supervised object detection (WSOD) has attracted extensive research attention due to its great flexibility of exploiting large-scale dataset with only image-level annotations for detector training. Despite its great advance in recent years, WSOD still suffers limited performance, which is far below that of fully supervised object d...More
PPT (Upload PPT)