A corpus-guided framework for robotic visual perception

摘要

We present a framework that produces sentence-level summa-rizations of videos containing complex human activities that can be implemented as part of the Robot Perception Control Unit (RPCU). This is done via: 1) detection of pertinent ob-jects in the scene: tools and direct-objects, 2) predicting ac-tions guided by a large lexical corpus and 3) generating the most likely sentence description of the video given the detec-tions. We pursue an active object detection approach by fo-cusing on regions of high optical flow. Next, an iterative EM strategy, guided by language, is used to predict the possible actions. Finally, we model the sentence generation process as a HMM optimization problem, combining visual detections and a trained language model to produce a readable descrip-tion of the video. Experimental results validate our approach and we discuss the implications of our approach to the RPCU in future applications.