Dataset Availability

The dataset used in this project is publicly available on Hugging Face. You can access and download it using the following link: https://huggingface.co/datasets/tsinghua-ee/AVUTBenchmark

Within the dataset, there are two video annotation JSON files:

• AV_Human_data.json – annotations created by human annotators

• AV_Gemini_data.json – annotations generated by the Gemini model

🤗 View Dataset on Hugging Face