Title :
Adaptive support-weight approach for correspondence search
Author :
Yoon, Kuk-Jin ; Kweon, In So
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., KAIST, Daejeon, South Korea
fDate :
4/1/2006 12:00:00 AM
Abstract :
We present a new window-based method for correspondence search using varying support-weights. We adjust the support-weights of the pixels in a given support window based on color similarity and geometric proximity to reduce the image ambiguity. Our method outperforms other local methods on standard stereo benchmarks.
Keywords :
image colour analysis; image resolution; search problems; adaptive support-weight approach; color similarity; correspondence search; geometric proximity; image ambiguity reduction; standard stereo benchmarks; support window; window-based method; Colored noise; Costs; Image analysis; Image segmentation; Layout; Performance evaluation; Pixel; Power measurement; Shape; Testing; 3D/stereo scene analysis.; Stereo; Algorithms; Artificial Intelligence; Biomimetics; Colorimetry; Feedback; Humans; Image Enhancement; Image Interpretation, Computer-Assisted; Imaging, Three-Dimensional; Pattern Recognition, Automated; Photogrammetry; Signal Processing, Computer-Assisted;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2006.70
