首页 /研究 /Finding the Correspondence of Audio-Visual Events by Object Manipulation
MANIPULATION

Finding the Correspondence of Audio-Visual Events by Object Manipulation

Kento Nishibori, Yoshinori Takeuchi, Tetsuya Matsumoto, Hiroaki Kudo, Noboru Ohnishi

发表年份
2008
引用次数
4
访问权限
开放获取

摘要

A human being understands the objects in the environment by integrating information obtained by the senses of sight, hearing and touch. In this integration, active manipulation of objects plays an important role. We propose a method for finding the correspondence of audio-visual events by manipulating an object. The method uses the general grouping rules in Gestalt psychology, i.e. “simultaneity” and “similarity” among motion command, sound onsets and motion of the object in images. In experiments, we used a microphone, a camera, and a robot which has a hand manipulator. The robot grasps an object like a bell and shakes it or grasps an object like a stick and beat a drum in a periodic, or non-periodic motion. Then the object emits periodical/non-periodical events. To create more realistic scenario, we put other event source (a metronome) in the environment. As a result, we had a success rate of 73.8 percent in finding the correspondence between audio-visual events (afferent signal) which are relating to robot motion (efferent signal).

关键词

Computer visionComputer scienceArtificial intelligenceObject (grammar)RobotMicrophoneMotion (physics)CommunicationSpeech recognitionPsychology

相关论文

查看 MANIPULATION 分类全部论文