CAT-Seg: Cost Aggregation for Open-Vocabulary Semantic Segmentation

Existing works on open-vocabulary semantic segmentation have utilized large-scale vision-language models, such as CLIP, to leverage their exceptional open-vocabulary recognition capabilities. However, the problem of transferring these capabilities learned from image-level supervision to the pixel-level task of segmentation and addressing arbitrary unseen categories at inference makes this task challenging. To address these issues, we aim to attentively relate objects within an image to given categories by leveraging relational information among class categories and visual semantics through aggregation, while also adapting the CLIP representations to the pixel-level task. However, we observe that direct optimization of the CLIP embeddings can harm its open-vocabulary capabilities. In this regard, we propose an alternative approach to optimize the image-text similarity map, i.e. the cost map, using a novel cost aggregation-based method. Our framework, namely CAT-Seg, achieves state-of-the-art performance across all benchmarks. We provide extensive ablation studies to validate our choices. Project page:

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Open Vocabulary Semantic Segmentation ADE20K-150 CAT-Seg mIoU 36.2 # 1
Open Vocabulary Semantic Segmentation ADE20K-847 CAT-Seg mIoU 14.1 # 1
Open Vocabulary Semantic Segmentation PASCAL Context-459 CAT-Seg mIoU 21.4 # 1
Open Vocabulary Semantic Segmentation PASCAL Context-59 CAT-Seg mIoU 61.5 # 1
Open Vocabulary Semantic Segmentation PascalVOC-20 CAT-Seg mIoU 97.1 # 1
Open Vocabulary Semantic Segmentation PascalVOC-20b CAT-Seg mIoU 81.4 # 1