Dataset · ropedia-ai
Egocentric multimodal experience data
ropedia-ai/xperience-10m
A large corpus of human action video, audio, and 3D data for robotics and embodied AI.
What's inside
Xperience-10M is a large-scale egocentric multimodal dataset of human experience. It features 3D, audio, and video modalities; tasks include video classification, image-to-text, depth estimation, and robotics. The language is English, and access is granted upon request and signing a DocuSign agreement. Metadata indicates a size of 1 to 10 million.
What you can do with it
- Train policies for embodied AI and robotics
- Fine-tune video and audio models on egocentric data
- Evaluate depth estimation and image-to-text models
- Tasks
- Video classification, Image → text, Depth estimation, Robotics
- Languages
- English
- Size
- 1M–10M rows
- Downloads this month
- 43K
- Likes
- 257
- Access
- on request on Hugging Face
Models trained on it
- cy0307/ropedia-xperience-10m-task-baselinesRobotics 248
- cy0307/ropedia-qwen3-omni-lora-128epRobotics 42
- cy0307/ropedia-qwen3-omni-lora-smokeRobotics 17
- tekwhisperer/Hemppimp 0
- solomonmc/Virra 0
- AMT21/ZIMFARM 0
- cy0307/ropedia-cosmos3-super-forward-dynamics-lora-128epRobotics 0
- cy0307/ropedia-xperience-10m-weights-resultsRobotics 0
How to load it
from datasets import load_dataset
ds = load_dataset("ropedia-ai/xperience-10m")