Learning by watching: extracting reusable task knowledge from visual observation of human performance

By Yasuo Kuniyoshi, Masayuki Inaba, Hirochika Inoue. 1994

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Combining self-supervised learning and imitation for vision-based rope manipulation (2017, background): “Many past works have proposed good solutions to learning from visual demonstrations [29], [30], [4].”
Zero-Shot Visual Imitation (2018, background): “A large body of work (Breazeal & Scassellati, 2002; Dillmann, 2004; Ikeuchi & Suehiro, 1994; Kuniyoshi et al., 1989; 1994; Yang et al., 2015) has sought to capture prior knowledge by manually pre-defining the state that must be ∗Denotes equal contribution. ar X iv :1 80 4.”
FlashMeta: a framework for inductive program synthesis (2015, background): “It has applications to a wide variety of domains, including robotics [20], software development [19], and biology [18].”
Learning from observation using primitives (2001, background): “such as trajectory learning [1], learning from observation [4, 5, 11, 15, 6, 10, 12, 13], postural primitives [21], and other techniques that decompose complex tasks or movements into smaller parts [2, 7, 16].”
A unifying computational framework for motor control and social interaction. (2003, methodology): “This method of action observation contrasts with previous methods of imitation learning that use several heuristic methods for storing features of movement patterns, for example, points of high curvature or discontinuity (Kuniyoshi et al. 1994; Wada et al. 1995; Miyamoto et al. 1996).”
Incremental learning of full body motion primitives and their sequencing through human motion observation (2012, background): “Robot skill learning through observation of human motion is an attractive paradigm for humanoid robots, and has received significant attention in the literature (Kuniyoshi et al. 1989; Kuniyoshi and Inoue 1994; Liu and Asada 1992; Dillmann et al. 1999).”
Evaluation of robot imitation attempts: comparison of the system's and the human's perspectives (2006, unknown): “The major directions of work have been using such behaviours with humans for adaptive learning [8, 12, 6, 3, 7] and also allowing the robots to be able to engage in richer, more natural social interactions [5, 11, 4].”
A syntactic approach to robot imitation learning using probabilistic activity grammars (2013, unknown): “Our approach shares the concept of imitation learning presented in the Handbook of Robotics (Chapter 59) [1], as well as in [2–5] where a robot learns a new task directly from human demonstration without the need of extensive reprogramming.”
Improved human–robot team performance through cross-training, an approach inspired by human team training practices (2015, background): “In Kuniyoshi et al. (1994), a robot extracts a description of the task by tracking human hand motions through a vision system.”
Humanoid Robot's Autonomous Acquisition of Proto-Symbols through Motion Segmentation (2006, background): “Kuniyoshi et al have proposed a learning framework for robots that can reproduce complex behavior through observing motion patterns of performers [4].”
Discovering optimal imitation strategies (2004, methodology): “The imitation mechanism was, then, aimed at a precise reproduction of a pre-specified sub-set of task features, such as hand–object actions (picking up a block, rotating the block) [9–11,17], state-actions (turn left, move forward) [6,7], the path followed by”
Task-Specific Generalization of Discrete and Periodic Dynamic Movement Primitives (2010, background): “Already, early research on learning from demonstration has stressed the importance of task segmentation and the extraction of meaningful action units [13]–[15].”
Acquisition of Probabilistic Behavior Decision Model based on the Interactive Teaching Method (2001, background): “This problem is caused by xed expression of relation between visual information and behaviors, as same as the teaching by showing which is Kuniyoshi's work[3].”
Exploration and apprenticeship learning in reinforcement learning (2005, background): “Other examples include Sammut et al. (1992); Kuniyoshi, Inaba & Inoue (1994); Demiris & Hayes (1994); Amit & Mataric (2002); and Pomerleau (1989), which apply supervised learning to learn a parameterized policy from the demonstrations.”