Reproduce “SPiKE: 3D Human Pose from Point Cloud Sequences” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Irene Ballester, Ondřej Peterka, Martin Kampel
SPiKE estimates 3D human poses from point-cloud sequences by extracting local spatial features within each timestamp and using a Transformer to relate them across time. This lets earlier frames provide context when a joint is occluded. On ITOP front-view, it reports 89.19% mAP, above DECA but below AdaPose; the evidence is limited to single-human depth-derived point clouds.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.