TS-CAM: Token Semantic Coupled Attention Map for Weakly Supervised Object Localization
This addresses the problem of incomplete object localization in weakly supervised learning for computer vision applications, representing a novel method rather than an incremental improvement.
The paper tackles the partial activation issue in weakly supervised object localization by proposing TS-CAM, which uses a visual transformer to capture long-range dependencies and re-allocates semantics, resulting in state-of-the-art performance with improvements of 7.1% on ILSVRC and 27.1% on CUB-200-2011 datasets.
Weakly supervised object localization (WSOL) is a challenging problem when given image category labels but requires to learn object localization models. Optimizing a convolutional neural network (CNN) for classification tends to activate local discriminative regions while ignoring complete object extent, causing the partial activation issue. In this paper, we argue that partial activation is caused by the intrinsic characteristics of CNN, where the convolution operations produce local receptive fields and experience difficulty to capture long-range feature dependency among pixels. We introduce the token semantic coupled attention map (TS-CAM) to take full advantage of the self-attention mechanism in visual transformer for long-range dependency extraction. TS-CAM first splits an image into a sequence of patch tokens for spatial embedding, which produce attention maps of long-range visual dependency to avoid partial activation. TS-CAM then re-allocates category-related semantics for patch tokens, enabling each of them to be aware of object categories. TS-CAM finally couples the patch tokens with the semantic-agnostic attention map to achieve semantic-aware localization. Experiments on the ILSVRC/CUB-200-2011 datasets show that TS-CAM outperforms its CNN-CAM counterparts by 7.1%/27.1% for WSOL, achieving state-of-the-art performance.