Self-Erasing Network for Integral Object Attention
This work addresses a key issue in weakly-supervised object localization for computer vision applications, representing an incremental improvement over existing adversarial erasing methods.
The paper tackled the problem of attention regions expanding to non-object areas in weakly-supervised object attention, introducing a Self-Erasing Network (SeeNet) that effectively highlights integral object regions with reduced background inclusion, as demonstrated by superior results on Pascal VOC.
Recently, adversarial erasing for weakly-supervised object attention has been deeply studied due to its capability in localizing integral object regions. However, such a strategy raises one key problem that attention regions will gradually expand to non-object regions as training iterations continue, which significantly decreases the quality of the produced attention maps. To tackle such an issue as well as promote the quality of object attention, we introduce a simple yet effective Self-Erasing Network (SeeNet) to prohibit attentions from spreading to unexpected background regions. In particular, SeeNet leverages two self-erasing strategies to encourage networks to use reliable object and background cues for learning to attention. In this way, integral object regions can be effectively highlighted without including much more background regions. To test the quality of the generated attention maps, we employ the mined object regions as heuristic cues for learning semantic segmentation models. Experiments on Pascal VOC well demonstrate the superiority of our SeeNet over other state-of-the-art methods.