facebookresearch/dinov3

PyTorch models for high-res vision tasks using self-supervised learning

VGGT is a neural network that predicts 3D scene attributes like camera parameters, depth, and point maps from images
facebookresearch/vggt
PyTorch computer vision toolkit with self-supervised learning, transformers, and detection models
alibaba/EasyCV
Open vision-language models that compete with closed-source alternatives using fully open data
facebookresearch/perception_models
video and image watermarking code for training and inference
facebookresearch/videoseal
Video models like SlowFast help computers spot actions and understand what’s happening in videos by combining slow and fast motion cues
facebookresearch/SlowFast
State-of-the-art object detection and segmentation library from Facebook AI Research
facebookresearch/detectron2
Collection of PyTorch image models, training scripts, and utilities
huggingface/pytorch-image-models
Big Vision: codebase for training large-scale vision models using Jax/Flax andTPUs
google-research/big_vision
data augmentations for audio, image, text, and video
facebookresearch/AugLy
tracks any pixel in video with transformer-based model, supports dense and long-term point tracking even through occlusion
facebookresearch/co-tracker