AI · AI model
EgoX
AI framework generating first-person egocentric videos from third-person video footage
EgoX is a research framework developed by KAIST AI and Seoul National University that synthesizes egocentric (first-person) video from a single exocentric (third-person) input. It lifts 2D footage into 3D point clouds to establish viewpoint priors, then refines the output using pretrained video diffusion models with custom geometry-guided attention. This allows researchers and creators to generate realistic POV shots, including hands and interaction details, from standard filmed scenes.
What it's for
- Synthesize first-person camera perspectives from third-person movie clips or sports footage
- Generate egocentric training data for robotics and embodied AI agents using existing standard video
- Create immersive POV scenes from external camera angles for VFX and digital content creation
Features
- Single exocentric video input processing
- 3D point cloud lifting to project egocentric spatial priors
- Width-wise and channel-wise latent concatenation to fuse viewpoints
- Geometry-guided self-attention to maintain spatial alignment and avoid visual artifacts
- Lightweight LoRA adaptation of pretrained video diffusion models
Pricing
Open source
Details
- Type
- AI model
- Category
- AI
- Topic
- AI Video Generation
- Open source
- Yes
- Platforms
- Linux, Self-hosted
- For
- Computer vision researchers, embodied AI engineers, and visual effects artists
- Tags
- video diffusion, egocentric, computer vision, point cloud, research paper
Why it was saved
- @evolving.ai: Allows creators and researchers to synthesize realistic first-person camera perspectives from existing third-person video clips.
Related
AI creative suite for generating and editing video, images, and animation
AI platform generating and editing video, image, and synchronized audio.
AI video and image generator known for cinematic motion and physics
AI-assisted multitrack video editor and text-to-video generator for creators and teams.
AI video and image generator with advanced motion control and physical simulation
Generative AI video model producing cinematic footage with native audio
Updated