Dataset · Mutonix
Video Clips with Script Annotations and Camera Motion
Mutonix/Vript
About 400k English video clips with detailed scene, shot, and camera motion descriptions.
What's inside
The dataset consists of 12,000 high-resolution videos sliced into approximately 400,000 clips. Annotation is done in screenwriting format: alongside the scene content, shot types (medium, close-up, etc.) and camera motion types (pan, tilt, and others) are specified.
What you can do with it
- Training text-to-video generation models with camera control
- Training video-language models to understand cinematic techniques
- Video question answering (VQA) and video classification tasks
- Tasks
- Video classification, Visual question answering, Text → video, Text → image, Image → video
- Languages
- English
- Size
- 100K–1M rows
- Downloads this month
- 37.3K
- Likes
- 27
Models trained on it
- HuggingFaceTB/SmolVLM2-500M-Video-InstructImage text → text 1.7M
- HuggingFaceTB/SmolVLM2-2.2B-InstructImage text → text 194.6K
- HuggingFaceTB/SmolVLM2-256M-Video-InstructImage text → text 44.7K
- HuggingFaceTB/SmolVLM2-2.2B-BaseImage text → text 2.4K
- mradermacher/SmolVLM2-2.2B-Instruct-i1-GGUFVideo text → text 1.1K
- mradermacher/SmolVLM2-256M-Video-Instruct-i1-GGUF 824
- mradermacher/SmolVLM2-500M-Video-Instruct-i1-GGUF 737
- mradermacher/SmolVLM2-2.2B-Instruct-GGUFVideo text → text 368
How to load it
from datasets import load_dataset
ds = load_dataset("Mutonix/Vript")