Biologically inspired detection for temporal texture motion
Bin Sun, Kaiyu Qin · 2011
Motion detection has always been a challenge for computer vision. A large variety of algorithms have been proposed for diverse form applications. Nevertheless, the traditional approaches are mainly based on brightness cues, or rather task-driven. Inspired by biological vision theory, in this paper, a novel second order motion detector is proposed for computer vision applications. The proposed implementation effectively utilizes the high-order motion information, rather than mere luminance pattern. Preliminary tests are conducted on both artificial and natural scenes. The results demonstrate that the proposed second order motion detector can capture effective information from a representative high motion instance, e.g., the temporal texture motion, and thorough analysis suggests that the bio-plausible exploiture may bring some new advantage to image processing practice.