Sam-3D-Body in C++: zero-Python runtime for full-body 3D pose and mesh from RGB
The SAM3DBody-cpp is a standalone C++ inference engine designed for full-body 3D pose estimation without Python dependencies. It processes BGR images to generate body pose parameters, camera translations, and optional 3D mesh data. The project includes pre-built models and various tools for exporting keypoints and rendering.
- ▪SAM3DBody-cpp operates entirely in C++ and does not require Python at runtime.
- ▪It utilizes ONNX Runtime to produce detailed body pose parameters and 3D mesh vertices from images.
- ▪The project provides a CSV exporter for keypoints and includes a variety of pre-built model files hosted on HuggingFace.
Opening excerpt (first ~120 words) tap to expand
SAM3DBody-cpp Standalone C++ inference engine for SAM-3D-Body — zero Python dependency at runtime. Takes a BGR image and produces per-person MHR body pose parameters, camera translation, and optionally full 3D mesh vertices + 70 body/hand keypoints, all via ONNX Runtime + ggml. Also includes Python frontends that call the compiled shared library via ctypes, and a CSV exporter for the 70 MHR keypoints. Models Pre-built ONNX / GGUF / LBS model files are hosted on HuggingFace: https://huggingface.co/AmmarkoV/SAM3DBody-cpp-onnx-models Download SAM3DBody-cpp-onnx-models.zip, extract it, and place the resulting onnx/ directory at the root of this repository: # Download and extract wget https://huggingface.co/AmmarkoV/SAM3DBody-cpp-onnx-models/resolve/main/SAM3DBody-cpp-onnx-models.zip unzip…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.