facebookresearch/sam2
segment anything in images and videos with sam2
View on index · View in 3D Map
Related repos
- video and image watermarking code for training and inference
- EdgeTAM: on-device SAM 2 variant for segmentation and tracking in videos, runs faster, processes 16 FPS on iPhone 15 Pro Max
- data augmentations for audio, image, text, and video
- State-of-the-art object detection and segmentation library from Facebook AI Research
- Library for video understanding research
- Reconstructs 3D scenes from 1000+ images in one forward pass
- Video models like SlowFast help computers spot actions and understand what’s happening in videos by combining slow and fast motion cues
- AI model for cutting out objects from images with simple prompts
- turns audio into realistic talking avatars
- tracks any pixel in video with transformer-based model, supports dense and long-term point tracking even through occlusion
