CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
In this paper, we address the stereo matching problem in the presence of reflections and translucency, where image formation can be modeled as the additive superposition of layers...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
Traditional approaches for surface reconstruction from range data require that the input data be either range images or unorganized sets of points. Since a large number of range s...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
1 We address the problem of vector-valued image regularization with variational methods and PDE's. From the study of existing formalisms, we propose a unifying framework base...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
During tracking, lens zoom acts as a gain between scene dynamics and fixation errors, providing a trade-off between maximising resolution and minimising tracking error. Using a li...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
We introduce the problem of repetitive nearest neighbor search in relevance feedback and propose an efficient search scheme for high dimensional feature spaces. Relevance feedback...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
This paper compares two methods for object localization from contours: shape context and chamfer matching of templates. In the light of our experiments, we suggest improvements to...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
Existing color constancy methods cannot handle both uniform colored surfaces and highly textured surfaces in a single integrated framework. Statistics-based methods require many s...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
In many applications of graphical models arising in computer vision, the hidden variables of interest are most naturally specified by continuous, non-Gaussian distributions. There...
CVPR
2003
IEEE
15 years 3 days ago
2003
IEEE
In this paper, we propose a Bayesian approach to image hallucination. Given a generic low resolution image, we hallucinate a high resolution image using a set of training images. ...