S001-Living room
Indoor
Movement + Shade
Content: Content: Walk in → Sit down
S006 - Roadside
Indoor
Feature Focus:Gait change
Content: Walking → stopping → looking at cell phone
S009-Living room
Indoor
Reflective Surface
Content: Dressing + Mirror
S0010 - Empty Room
Indoor
Large movement
Content: Dance Moves
S0013 - Night Corridor
Indoor
Walking + Dimming
Content: Walking + Dimming Low Light
S0013 - Night Corridor
Indoor
Walking + Dimming
Content: Walking + Dimming Low Light
Core Value
Provides real-world synchronized multi-view videos to solve key challenges in generative vision AI
Spatial consistency across viewpoints
Natural motion transfer between perspectives
Occlusion handling in dynamic scenes
Physics-aware rendering with real-world lighting
Applications
3D Digital Humans
Scene Re-rendering
VR/AR Content Creation
Face swapping and scene editing
Why This Dataset?
True-to-life Capture
- Real-world lighting/shadows (vs. synthetic data)
- Professional camera movements comparable to film production
Structured Challenges
- Progressive difficulty from single-person to multi-person scenes
- Deliberate occlusion designs (e.g., S008 two-person crossing)
Production-Ready Format
- Frame-aligned MP4 files with EXIF metadata
- Pre-computed camera intrinsics/extrinsics
About maadaa.ai
Founded in 2015, maadaa.ai is a pioneering AI data service provider specializing in multimodal data solutions for generative AI development. We deliver end-to-end data services covering text, voice, image, and video datatypes – the core fuel for training and refining generative models.
Our Generative AI Data Solution includes:
ꔷ High-quality dataset collection & annotation tailored for LLMs and diffusion modelsꔷ Scenario-based human feedback (RLHF/RLAIF) to enhance model alignment
ꔷ One-stop data management through our MaidX platform for streamlined model training
Get Access to the Dataset