#9090th of 126 in ML frameworks and librariesFor science, health and research, science
deeplabcut
Markerless animal pose estimation toolbox with GUI and Python API
- Stars
- 5.8k
- License
- LGPL-3.0
- Last commit
- Sep 2026
- Last release
- Sep 2026
- Language
- Python
5,536 on 3 Apr 20265,810 on 30 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
DeepLabCut trains and runs pose estimation models that track user-labeled body parts of animals or objects in video. It offers a GUI and Python API, a PyTorch engine (v3.0+) with ResNet, HRNet and RTMPose backbones, and a legacy TensorFlow engine slated for deprecation in 3.2. It also ships SuperAnimal-Quadruped and SuperAnimal-TopViewMouse pretrained models, which are research-use only.
Who it is for: Life scientists and vision researchers tracking animal behavior in video
Strengths
- Works on any animal or object you can label; multi-animal and 3D pose supported
- Pretrained SuperAnimal models with published mAP tables on held-out datasets
- Both GUI and Python API; pip install with optional [gui] and [tf] extras
- Colab demo notebook and Nature Protocols workflow guide
Weaknesses
- SuperAnimal models are non-commercial, research use only
- TensorFlow backend is planned for deprecation in version 3.2
- Requires Python 3.10+ and a manual PyTorch/CUDA install for GPU
- Needs manual frame labeling to train on new subjects
What it needs
- GPU optional
- Docker
- Needs PyTorch, TensorFlow (optional legacy engine)
- Models: ResNet, HRNet, RTMPose, DLCRNet, MobileNetV2
Quick answers
Is deeplabcut open source?
Yes. Its code is on GitHub under the LGPL-3.0 license.
Can you run deeplabcut with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does deeplabcut need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in ML frameworks and libraries
See all 126| Rank | Project | Score |
|---|---|---|
| 1 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 95 out of 100 |
| 2 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |