MSR-VTT

The MSRVTT dataset is a benchmark used to evaluate video-text retrieval (VTR) systems by providing a collection of videos paired with natural language queries.

Papers using MSR-VTT (3)

Low-rank HOCA: Efficient High-order Cross-modal Attention For Video Captioning2019 · 14 citesDelving Deeper: Hierarchical Visual Perception for Robust Video-Text Retrieval2026Leveraging Modality Tags for Enhanced Cross-Modal Video Retrieval2025