Katha AI, IIIT Hyderabad
Website under construction

The Katha-AI research group is led by Dr. Makarand Tapaswi at the Center for Visual Information Technology (CVIT), IIIT Hyderabad, India.

Here are some of our group's major projects:

• [CVPR 2025] VELOCITI: Benchmarking Video-Language Compositional Reasoning with Strict Entailment
• [TMLR 2025] No Detail Left Behind: Revisiting Self-Retrieval for Fine-Grained Image Captioning
• [WACV 2025] Seeing Eye to AI: Comparing Human Gaze and Model Attention in Video Memorability
• [ECCVW 2024] Detect, Describe, Discriminate: Moving Beyond VQA for MLLM Evaluation
• [CVPR 2024] MICap: A Unified Model for Identity-aware Movie Descriptions
• [CVPR 2024] "Previously on ..." From Recaps to Story Summarization
• [CVPR 2023] How you feelin'? Learning Emotions and Mental States in Movie Scenes