358 search results - page 45 / 72 » Scene-Consistent Detection of Feature Points in Video Sequen... |
ICCV
2011
IEEE
12 years 7 months ago
2011
IEEE
Fitting statistical 2D and 3D shape models to images is necessary for a variety of tasks, such as video editing and face recognition. Much progress has been made on local fitting...
DAGM
2009
Springer
13 years 5 months ago
2009
Springer
Abstract. Guided Kanade-Lucas-Tomasi (GKLT) tracking is a suitable way to incorporate knowledge about camera parameters into the standard KLT tracking approach for feature tracking...
ICMCS
2005
IEEE
14 years 1 months ago
2005
IEEE
The lack of eye contact in video conference degrades the user’s experience. This problem has been known and studied for many years. There are hardware-based solutions to the eye...
ECCV
2006
Springer
14 years 9 months ago
2006
Springer
Abstract Where feature points are used in real-time frame-rate applications, a high-speed feature detector is necessary. Feature detectors such as SIFT (DoG), Harris and SUSAN are ...
CVPR
2000
IEEE
13 years 12 months ago
2000
IEEE
This paper describes a fast and robust approach to recovering structure and motion from video frames. It rst describes a robust recursive factorization method for ane projection...