This Volume 2 of the conference proceedings contains 95 papers. Topics discussed include robot localization and navigation, video, learning and statistical methods, vision for graphics, reflectance modeling, calibrati...
详细信息
This Volume 2 of the conference proceedings contains 95 papers. Topics discussed include robot localization and navigation, video, learning and statistical methods, vision for graphics, reflectance modeling, calibration and structure from motion, tracking, recognition, three dimensional reconstruction, fingerprint recognition and registration and alignment.
We propose a new near-real time technique for 3D face pose tracking from a monocular image sequence obtained from a n uncalibrated camera. The basic idea behind our approach is that instead of treating 2D face detecti...
详细信息
A system is presented for automatic classification of the atmosphere in motion pictures scenes. This classification is based solely on the extraction of approximate illumination information from single frames. By usin...
详细信息
A system is presented for automatic classification of the atmosphere in motion pictures scenes. This classification is based solely on the extraction of approximate illumination information from single frames. By using well defined film grammar rules regarding scene lighting, it generates high level semantic descriptors for frames and scenes. These illumination descriptors could be further combined with other descriptors like motion and sound, to generate a complete automatic semantic analysis system of video media.
The Gaussian kernel has played a central role in multi-scale methods for feature extraction and matching. In this paper, a method for shaping the filter using the local image structure is presented. We propose an opti...
详细信息
The Gaussian kernel has played a central role in multi-scale methods for feature extraction and matching. In this paper, a method for shaping the filter using the local image structure is presented. We propose an optimization formulation that densely estimates the filter's affine parameters by minimizing an objective constructed from differential feature responses and seek iterative, approximate solutions. A consequence of shaping the filters is affine invariance of the differential feature vector and it is shown that the shaped responses improve recognition performance.
A novel region-based progressive stereo matching algorithm is presented. It combines the strengths of previous region-based and progressive approaches. The progressive framework avoids the time consuming global optimi...
详细信息
A novel region-based progressive stereo matching algorithm is presented. It combines the strengths of previous region-based and progressive approaches. The progressive framework avoids the time consuming global optimization, while the inherent problem, the sensitivity to early wrong decisions, is significantly alleviated via the region-based representation. A growing-like process matches the regions progressively using a global best-first strategy based on a cost function integrating disparity smoothness and visibility constraint. The performance on standard evaluation platform using various real images shows that the algorithm is among the state-of-the-art both in accuracy and efficiency.
暂无评论