Siggraph 2005 logo
Conference Exhibition Call For Participation Presenters Media Registration

emerging technologies

Haptic Video

Professional skills are typically taught with conventional materials such as videotapes that demonstrate expert procedures. But instruction is most effective when the haptic sense is proactive.

Art and Science

Since ancient times, artisans' skills have been handed down through training and guidance. Usually, the process requires a long-term commitment as an apprentice or many years of trial-and-error learning. Trainees can not obtain precisely the same skill-specific information that the experts apply during their work. Mental skills, on the other hand, can be presented precisely through language.

Recent technological developments have introduced a new capability: recording and reproducing movement and force information. In Haptic Video, this technology transmits physical operations with the same precision as language. By recording and reproducing an expert's procedures proactively, it demonstrates how humans can use technology to improve their skills in a wide variety of professional and aesthetic fields, from medicine to calligraphy to etiquette.


Haptic Video provides a precise and proactive approach to transmitting physical skills. By recording the working environment as well as the movement and force of an expert instructor, the goals are to develop an archiving system so that all pertinent skill information can be reproduced dynamically, clarify the meaning of proactivity for "active touch," verify the system's effectiveness as a training device, and demonstrate how haptic devices can be used for substantial improvement of existing skills.


1. In the recording phase, position and applied force are recorded as the expert works. This information is transformed into position and impedance information, and archived in a database that can be dynamically interpolated. In the presentation phase, the impedance information is presented to the trainee along the trajectory direction, and virtual fixtures, which are like walls with elasticity, are presented orthogonal to the trajectory direction at the same time. As a result, the trainee tries to cancel the force that the expert exerted, and duplicate the desired force proactively.

2. The working environment near the expert's hand is recorded from the expert's point of view. Then, in the presenting phase, the playback speed of the video is dynamically changed according to location information obtained from the trainee's proactive operation. As a result, the trainee can observe the changes near the expert's hand as an image corresponding to the trainee's movements. In this way, both vision and tactile information corresponding to an expert's skill, which until now could only be obtained passively, can be transmitted proactively.


Common linguistic expressions such as "press softer" and "be more rhythmical" are inherently vague. Conversely, exact verbal instructions such as "apply three kilograms" or "twist twelve degrees" cannot be realized precisely. Neither are sufficient to transmit physical information, which is key to acquiring complex skills. When trainees proactively reproduce the operation of an expert, the expert's force and images can be perceived, and skill can be transmitted not by passive interpretation but by an immediate and proactive expression of the haptic sense, with great accuracy. Future projects will clarify what is important for acquisition and improvement of physical skills by experimenting with transmission techniques.


Satoshi Saga
Tachi Laboratory, The University of Tokyo
satoshi_saga (at)


Kevin Vlack
Hiroyuki Kajimoto
Susumu Tachi
Tachi Laboratory, The University of Tokyo