Keypoints into the Future

Self-Supervised Correspondence in Model-Based RL

Project Video

5 minute video overview for the Conference on Robot Learning (CoRL 2020)

Publication

Lucas Manuelli, Yunzhu Li, Pete Florence, Russ Tedrake. Keypoints into the Future: Self-Supervised Correspondence in Model-Based Reinforcement Learning. Conference on Robot Learning, 2020. [paper]

Extended Video (Hardware Results)

Supplementary Material

KITF_appendix.pdf

Code (coming soon)