zetyquickly/DensePoseFnL

Convert your videos to densepose and use it on MagicAnimate

OpenMMLab Pose Estimation Toolbox and Benchmark.

1K resolution vision transformers pretrained on 1B human images.

GTR: Gated Token Recurrence for Efficient Dense Prediction

OpenPose: Real-time multi-person keypoint detection library for body, face, hands, and foot estimation

VPoser: Variational Human Pose Prior

DETRPose: Real-Time End-to-End Multi-Person Pose Estimation via Modified Transformer Decoder and Novel Denoising Keypoints [ECCV 2026]

The repository provides code for running inference with the SAM 3D Body Model (3DB), links for downloading the trained model checkpoints and datasets, and example notebooks that show how to use the model.

Train a stacked hourglass deep neural network for human pose estimation on the COCO 2017 dataset.