Foundation Pose
Software
Extracting 6D object pose sequences from human video demonstrations for goal conditioning
"We use Foundation Pose and SAM to extract the sequence of goal poses to track, and then the RL policy tracks these goals one by one in a 60 Hz control loop."