Maddog241/mvinverse
Predict scene properties from image sequences
View on index · View in 3D Map
Related repos
- VGGT is a neural network that predicts 3D scene attributes like camera parameters, depth, and point maps from images
- Reconstructs 4D human scenes from videos
- Turn images into cinematic videos with 33 facial expressions and 400+ motion combinations
- Generate video and 3D scenes via unified world modeling
- Trains models from unlabeled images using foundation models
- Generates 4K images with FLUX models
- control diffusion models with extra conditions for image generation
- Visual regression testing for Storybook and web pages
- Restores images and videos through super-resolution
- bitwise autoregressive model for fast high-res text-to-image generation
