| ## Description | |
| 100,000 Sets: Egocentric Procedural Reasoning Video Dataset includes first-person perspective, multitasking interactive videos covering tasks such as cooking, crafting, and sports. Annotation content comprises both overall descriptions and detailed descriptions of actions. | |
| For more details, please refer to the link: https://www.nexdata.ai/datasets/computervision?source=Huggingface | |
| ## Specifications | |
| ### Data Size | |
| 100000 sets | |
| ### Data Format | |
| The video is in common formats such as MP4 and labeled as JSON format | |
| ### Data Type | |
| First person character multitasking interactive video, with tasks covering cooking, crafting, sports, etc | |
| ### Annotation Content | |
| Including overall description and intensive description of actions | |
| ## Licensing Information | |
| Commercial License |