Abstract: Utilizing a single vision sensor for learning from human demonstration (LfD) planning offers numerous benefits. However, the accurate identification of key task constraints for motion ...