%0 Generic %A Vecerik, Mel %A Doersch, Carl %A Yang, Yi %A Davchev, Todor %A Aytar, Yusuf %A Zhou, Guangyao %A Hadsell, Raia %A Agapito, Lourdes %A Scholz, Jon %C Yokohama, Japan %D 2024 %F discovery:10196975 %I IEEE %K Training, Visualization, Tracking, Stacking, Production facilities, Planning, Task analysis %P 5397-5403 %T RoboTAP: Tracking Arbitrary Points for Few-Shot Visual Imitation %U https://discovery.ucl.ac.uk/id/eprint/10196975/ %V 2024 %X For robots to be useful outside labs and specialized factories we need a way to teach them new useful behaviors quickly. Current approaches lack either the generality to onboard new tasks without task-specific engineering, or else lack the data-efficiency to do so in an amount of time that enables practical use. In this work we explore dense tracking as a representational vehicle to allow faster and more general learning from demonstration. Our approach utilizes Track-Any-Point (TAP) models to isolate the relevant motion in a demonstration, and parameterize a low-level controller to reproduce this motion across changes in the scene configuration. We show this results in robust robot policies that can solve complex object-arrangement tasks such as shape-matching, stacking, and even full path-following tasks such as applying glue and sticking objects together, all from demonstrations that can be collected in minutes. %Z This version is the author-accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.