Reproduce “Learning Human Sports Motion Feasibility via Quantile Regression on Tracking Data” 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
Marc Gutiérrez-Pérez, Antonio Agudo
The six-input MLP predicts time-to-arrival at hypothetical targets from tracking data, learning the first percentile of observed times using speed, acceleration, and target geometry relative to heading. On held-out World Cup matches, its mean absolute error is 2.596 ± 0.005 seconds, versus 3.065 seconds for a learned reachable-hull baseline. This is a conditional lower envelope, not an absolute physical minimum.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.