vanoracai/Exploiting-Spatial-temporal-Relationships-for-3D-Pose-Estimation-via-Graph-Convolutional-Networks

This project is a reference implementation for an ICCV 2019 paper. While it was academically significant at the time for introducing GCNs to spatial-temporal 3D pose lifting, it is now functionally obsolete in the fast-moving computer vision landscape. Quantitatively, with 118 stars and 0 velocity over 6 years, it serves as a historical archive rather than an active development tool. Defensibility is low because the 'lifting' problem (2D to 3D) has since been more effectively solved by Transformer-based architectures (e.g., PoseFormer, MixSTE) and diffusion models which handle long-range temporal dependencies better than early GCNs. Frontier labs like Google (MediaPipe) and Apple (Vision framework) already provide production-grade 3D pose estimation as commodity APIs, leaving little room for niche academic implementations to survive outside of specific research citations. Competitive projects like OpenMMLab's MMPose offer much more robust, maintained, and performant versions of similar algorithms, making this specific repository a 'reproducibility' artifact rather than a viable product or foundational library.