Top-K Pooling with Patch Contrastive Learning for Weakly-Supervised Semantic Segmentation
This work addresses the challenge of generating reliable pseudo labels for semantic segmentation using only image-level labels, which is incremental as it builds on existing Vision Transformer-based methods.
The paper tackles the problem of inaccurate patch classification in weakly-supervised semantic segmentation by introducing a top-K pooling layer and patch contrastive learning, resulting in state-of-the-art performance on the PASCAL VOC 2012 dataset.
Weakly Supervised Semantic Segmentation (WSSS) using only image-level labels has gained significant attention due to cost-effectiveness. Recently, Vision Transformer (ViT) based methods without class activation map (CAM) have shown greater capability in generating reliable pseudo labels than previous methods using CAM. However, the current ViT-based methods utilize max pooling to select the patch with the highest prediction score to map the patch-level classification to the image-level one, which may affect the quality of pseudo labels due to the inaccurate classification of the patches. In this paper, we introduce a novel ViT-based WSSS method named top-K pooling with patch contrastive learning (TKP-PCL), which employs a top-K pooling layer to alleviate the limitations of previous max pooling selection. A patch contrastive error (PCE) is also proposed to enhance the patch embeddings to further improve the final results. The experimental results show that our approach is very efficient and outperforms other state-of-the-art WSSS methods on the PASCAL VOC 2012 dataset.