Weakly-Supervised Semantic Segmentation via Sub-category Exploration
This addresses the challenge of obtaining accurate object localization in semantic segmentation without dense annotations, but it is incremental as it builds on existing weakly-supervised methods.
The paper tackles the problem of weakly-supervised semantic segmentation with image-level annotations, where existing methods focus on discriminative object parts, by proposing a self-supervised approach using sub-category exploration to improve response maps, resulting in favorable performance against state-of-the-art methods.
Existing weakly-supervised semantic segmentation methods using image-level annotations typically rely on initial responses to locate object regions. However, such response maps generated by the classification network usually focus on discriminative object parts, due to the fact that the network does not need the entire object for optimizing the objective function. To enforce the network to pay attention to other parts of an object, we propose a simple yet effective approach that introduces a self-supervised task by exploiting the sub-category information. Specifically, we perform clustering on image features to generate pseudo sub-categories labels within each annotated parent class, and construct a sub-category objective to assign the network to a more challenging task. By iteratively clustering image features, the training process does not limit itself to the most discriminative object parts, hence improving the quality of the response maps. We conduct extensive analysis to validate the proposed method and show that our approach performs favorably against the state-of-the-art approaches.