QA-CLIMS: Question-Answer Cross Language Image Matching for Weakly Supervised Semantic Segmentation

Keyphrases

• Question Answering 100%
• Image Matching 100%
• Object Region 100%
• Weakly Supervised Semantic Segmentation 100%
• Language Image 100%
• Background Region 100%
• Cross-language 100%
• Class Activation Map 66%
• False Activation 66%
• Popular 33%
• Text-dependent 33%
• Foreground Region 33%
• Activation Map 33%
• Text-to-image 33%
• Query Image 33%
• MS COCO 33%
• COCO Dataset 33%
• Contrastive Learning 33%
• PASCAL VOC 33%
• Object Localization 33%
• Vision-language 33%
• Map Method 33%
• Visual Question Answering 33%
• Question Answering Model 33%
• Prompt Engineering 33%
• Contrastive Network 33%
• Textual Information 33%
• Open-vocabulary 33%
• Image-level Label 33%
• Foundation Models 33%

Computer Science

• Answer Question 100%
• Image Matching 100%
• Weakly-Supervised Semantic Segmentation 100%
• Background Region 100%
• Contrastive Learning 33%
• Visual Question Answering 33%
• Prompt Engineering 33%
• Textual Information 33%