SLV: Spatial Likelihood Voting for Weakly Supervised Object Detection
This addresses the challenge of accurate object localization in WSOD for computer vision applications, representing an incremental improvement over existing MIL-based methods.
The paper tackles the problem of weakly supervised object detection (WSOD) where methods often localize only discriminative parts rather than whole objects, proposing a spatial likelihood voting (SLV) module to improve localization without bounding box annotations. It achieves superior performance on PASCAL VOC 2007 and 2012 datasets.
Based on the framework of multiple instance learning (MIL), tremendous works have promoted the advances of weakly supervised object detection (WSOD). However, most MIL-based methods tend to localize instances to their discriminative parts instead of the whole content. In this paper, we propose a spatial likelihood voting (SLV) module to converge the proposal localizing process without any bounding box annotations. Specifically, all region proposals in a given image play the role of voters every iteration during training, voting for the likelihood of each category in spatial dimensions. After dilating alignment on the area with large likelihood values, the voting results are regularized as bounding boxes, being used for the final classification and localization. Based on SLV, we further propose an end-to-end training framework for multi-task learning. The classification and localization tasks promote each other, which further improves the detection performance. Extensive experiments on the PASCAL VOC 2007 and 2012 datasets demonstrate the superior performance of SLV.