Improve CAM with Auto-adapted Segmentation and Co-supervised Augmentation
This work improves object localization in computer vision for applications like image analysis, though it is incremental as it builds on existing CAM methods.
The paper tackled the problem of improving Class Activation Maps (CAM) for Weakly Supervised Object Localization by addressing background content, proposing a confidence segmentation module and co-supervised augmentation to enhance localization. The result was state-of-the-art performance with Top-1 localization errors of 37.69% on CUB-200 and 48.81% on ILSVRC datasets.
Weakly Supervised Object Localization (WSOL) methods generate both classification and localization results by learning from only image category labels. Previous methods usually utilize class activation map (CAM) to obtain target object regions. However, most of them only focus on improving foreground object parts in CAM, but ignore the important effect of its background contents. In this paper, we propose a confidence segmentation (ConfSeg) module that builds confidence score for each pixel in CAM without introducing additional hyper-parameters. The generated sample-specific confidence mask is able to indicate the extent of determination for each pixel in CAM, and further supervises additional CAM extended from internal feature maps. Besides, we introduce Co-supervised Augmentation (CoAug) module to capture feature-level representation for foreground and background parts in CAM separately. Then a metric loss is applied at batch sample level to augment distinguish ability of our model, which helps a lot to localize more related object parts. Our final model, CSoA, combines the two modules and achieves superior performance, e.g. $37.69\%$ and $48.81\%$ Top-1 localization error on CUB-200 and ILSVRC datasets, respectively, which outperforms all previous methods and becomes the new state-of-the-art.