An Investigation of Methods of First-Person Human Activity Recognition for Robot

First-person activity recognition, a subcategory of video-based human action recognition, has become an active research topic in computer vision. Action recognition systems is to automatically analyze the ongoing events. Unlike the traditional action recognition systems where videos or image sequences were recorded with out of reach cameras, a first-person activity recognition systems uses videos or image sequences that are recorded with cameras which individuals wear. Based on the emerging research in recent years, it is a time to review the existing approaches to first-person activity recognition and inspire more ideas to advance the field of research. We present a detailed overview of the state-of-the art method and covers all aspects of the general framework of first-person activity recognition. All related databases are introduced with details. Finally, we suggest views and thoughts for future researches

کلیدواژه ها:

Computer Vision ، Motion Recognition ، Human Activity Recognition ، Motion Analysis