First-Person Vision

被引:77
作者
Kanade, Takeo [1 ]
Hebert, Martial [1 ]
机构
[1] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA
基金
美国国家科学基金会;
关键词
Computer vision; eye tracking; object recognition; quality of life technologies; RECOGNITION; TRACKING; MOVEMENT;
D O I
10.1109/JPROC.2012.2200554
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
For understanding the behavior, intent, and environment of a person, the surveillance metaphor is traditional; that is, install cameras and observe the subject, and his/her interaction with other people and the environment. Instead, we argue that the first-person vision (FPV), which senses the environment and the subject's activities from a wearable sensor, is more advantageous with images about the subject's environment as taken from his/her view points, and with readily available information about head motion and gaze through eye tracking. In this paper, we review key research challenges that need to be addressed to develop such FPV systems, and describe our ongoing work to address them using examples from our prototype systems.
引用
收藏
页码:2442 / 2453
页数:12
相关论文
共 81 条
[1]  
Aghazadeh O., 2011, 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), P3297, DOI 10.1109/CVPR.2011.5995731
[2]  
Agrawal M., 2005, P IEEE WORKSH MOT VI, P207
[3]  
ALOIMONOS J, 1987, INT J COMPUT VISION, V1, P333
[4]  
Aloimonos J., 1990, Proceedings. 10th International Conference on Pattern Recognition (Cat. No.90CH2898-5), P346, DOI 10.1109/ICPR.1990.118128
[5]  
[Anonymous], 2006, P 2006 S EYE TRACKIN, DOI DOI 10.1145/1117309.1117350
[6]  
[Anonymous], 2011, P WACV
[7]  
[Anonymous], IEEE C COMP VIS PATT
[8]  
[Anonymous], CMU KITCH DAT SET
[9]  
[Anonymous], 2009, P ADV NEUR INF PROC
[10]  
[Anonymous], 2008, P 2 INT CONV REH ASS