Reproduce “Grasping the Future: Joint 3D Hand-Object Forecasting From Internet Videos” 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
Rustin Soraki, Matthew Wallingford, Ryan Aby, Brandon Chee, Jason Ren, Ali Farhadi, Homanga Bharadhwaj, Roozbeh Mottaghi
Grasping-the-Future predicts hand and object motion together, including their changing grasp relationship, across forecasting and other tasks defined by which trajectory information is observed. It trains one generative model on annotated and reconstructed video trajectories. On 331 unseen windows, its 20-sample energy score improves over released methods, but this does not establish better single forecasts or equal-input comparisons.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.