4245 search results - page 785 / 849 » Input Modeling |
163
click to vote
MICCAI
2004
Springer
16 years 6 months ago
2004
Springer
In this paper, we present a novel method for intra-operative registration directly from monocular endoscopic images. This technique has the potential to provide a more accurate su...
185
click to vote
CVPR
1998
IEEE
16 years 8 months ago
1998
IEEE
A new approach to the recognition of temporal behaviors and activities is presented. The fundamental idea, inspired by work in speech recognition, is to divide the inference probl...
211
click to vote
ECCV
2008
Springer
16 years 8 months ago
2008
Springer
We propose an algorithm for semantic segmentation based on 3D point clouds derived from ego-motion. We motivate five simple cues designed to model specific patterns of motion and 3...
175
click to vote
ECCV
2006
Springer
16 years 8 months ago
2006
Springer
Abstract. In spite of over two decades of intense research, illumination and pose invariance remain prohibitively challenging aspects of face recognition for most practical applica...
175
click to vote
ECCV
2006
Springer
16 years 8 months ago
2006
Springer
Using the variational approaches to estimate optical flow between two frames, the flow discontinuities between different motion fields are usually not distinguished even when an an...