A Dual-Network Progressive Approach to Weakly Supervised Object Detection

Xuanyi Dong, Deyu Meng, Fan Ma, Yi Yang
2017 Proceedings of the 2017 ACM on Multimedia Conference - MM '17  
A major challenge that arises in Weakly Supervised Object Detection (WSOD) is that only image-level labels are available, whereas WSOD trains instance-level object detectors. A typical approach to WSOD is to 1) generate a series of region proposals for each image and assign the image-level label to all the proposals in that image; 2) train a classi er using all the proposals; and 3) use the classi er to select proposals with high con dence scores as the positive instances for another round of
more » ... aining. In this way, the image-level labels are iteratively transferred to instance-level labels. We aim to resolve the following two fundamental problems within this paradigm. First, existing proposal generation algorithms are not yet robust, thus the object proposals are o en inaccurate. Second, the selected positive instances are sometimes noisy and unreliable, which hinders the training at subsequent iterations. We adopt two separate neural networks, one to focus on each problem, to be er utilize the speci c characteristic of region proposal re nement and positive instance selection. Further, to leverage the mutual bene ts of the two tasks, the two neural networks are jointly trained and reinforced iteratively in a progressive manner, starting with easy and reliable instances and then gradually incorporating di cult ones at a later stage when the selection classi er is more robust. Extensive experiments on the PASCAL VOC dataset show that our method achieves state-of-the-art performance.
doi:10.1145/3123266.3123455 dblp:conf/mm/DongMMY17 fatcat:uybobtsuqvg5bnv5ajqcqrfot4