Energy optimization method based on message passing mechanism applied for stereo-vision problems
Keywords:
stereo-vision, graph algorithms, energy minimization, message passingAbstract
In the article the scanline energy optimization algorithm, its characteristics and implementation method with the help of message passing mechanism is considered. The method of searching for energy function minimum and the method of searching for the argument at which this minimum occurs have been observed in the review.References
Wang, L. Global stereo matching leveraged by sparse ground control points / L. Wang, R. Yang // CVPR. – 2011. – P. 3033–3040.
Scharstein, D. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms / D. Scharstein, R. Szeliski // IJCV. – 2002. – Vol. 47, no. 1–3. – P. 7–42.
Hirschmüller, H. Evaluation of Cost Functions for Stereo Matching / H. Hirschmüller, D. Scharstein // IEEE Conference on Computer Vision and Pattern Recognition. – 2007. – Vol. 1. – P. 1–8.
Birchfield, S. Depth discontinuities by pixel topixel stereo / S. Birchfield, C. Tomasi // IJCV. – 1999. – Vol. 35, no. 3. – P. 269–293.
Ohta, Y. Stereo by intra- and inter-scanline search / Y. Ohta and T. Kanade // TPAMI. – 1985. – Vol. 7, no. 2. – P. 139–154.
Cox, I. A maximum likelihood stereo algorithm / I. Cox, S. Hingorani, S. Rao, and B. Maggs // Computer Vision, Graphics and Image Processing. – 1996. – Vol. 63, no. 3. – P. 542–567.
Оптимизация энергии в задачах компьютерного зрения и алгоритмы на графах, лекция 1 [Электронный ресурс]. – http://www.lektorium.tv/lecture/?id=12997
