license: mit
size_categories:
- 1K<n<10K
configs:
- config_name: CaReBench
data_files:
- split: test
path: json/metadata.json
CaReBench: A Fine-grained Benchmark for Video Captioning and Retrieval
Yifan Xu, Xinhao Li, Yichun Yang, Desen Meng, Rui Huang, Limin Wang
๐ Introduction
๐ CaReBench is a fine-grained benchmark comprising 1,000 high-quality videos with detailed human-annotated captions, including manually separated spatial and temporal descriptions for independent spatiotemporal bias evaluation.
๐ ReBias and CapST Metrics are designed specifically for retrieval and captioning tasks, providing a comprehensive evaluation framework for spatiotemporal understanding in video-language models.
โก CaRe: A Unified Baseline for fine-grained video retrieval and captioning, achieving competitive performance through two-stage Supervised Fine-Tuning (SFT). CaRe excels in both generating detailed video descriptions and extracting robust video features.
๐ State-of-the-art performance on both detailed video captioning and fine-grained video retrieval. CaRe outperforms CLIP-based retrieval models and popular MLLMs in captioning tasks.
