Improved Census Transforms for Resource-Optimized Stereo Vision Speaker : Kai-Wen, Weng Author: Wade S. Fife, James K. Archibald Good afternoon, ladies and gentlemen. I’m Kai-Wen, Weng. Today, I’m going to give a talk on the Improved Census Transforms for Resource-Optimized Stereo Vision. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 23, NO. 1, JANUARY 2013
Outline Introduction Proposed Algorithm Experimental Results Sparse Census Transform Generalized Census Transform Experimental Results Conclusion First. I'll introduce the census transform of this paper. Next I’ll explain proposed algorithm on this paper At last, I’d like to highlight some of the experimental results and conclusions from this study. //// Next I’ll go on to outline the related work in this study.
Introduction
Introduction Computation costs algorithm: SAD (Sum of Absolute Differences) SSD (Sum of Squared Differences) NCC (Normalized Cross Correlation) CT (Census Transform) Reference Target Disparity
Introduction Census Transform : Color Gradient
Introduction After aggregation step: Census on colors Census on gradients
Introduction Sparse census[6] : Half of the bits The computation costs for the hamming distances are quite large. X [6] C. Zinner, M. Humenberger, K. Ambrosch, and W. Kubinger, “An optimized software-based implementation of a census-based stereo matching algorithm,” in Proc. 4th ISVC, 2008, pp. 216–227.
Introduction Mini-census[8] : X [8] N.-C. Chang, T.-H. Tsai, B.-H. Hsu, Y.-C. Chen, and T.-S. Chang,“Algorithm and architecture of disparity estimation with mini-census adaptive support weight,” IEEE Trans. Circuits Syst. Video Technol., vol. 20, no. 6, pp. 792–805, Jun. 2010.
Introduction Mini-census[8] : Mini-census adaptive support weight [8] N.-C. Chang, T.-H. Tsai, B.-H. Hsu, Y.-C. Chen, and T.-S. Chang,“Algorithm and architecture of disparity estimation with mini-census adaptive support weight,” IEEE Trans. Circuits Syst. Video Technol., vol. 20, no. 6, pp. 792–805, Jun. 2010.
Introduction The challenges: It is critical to… The enormous amount of computation required to identify the corresponding points in the images. It is critical to… maximize the accuracy and throughput of the stereo system while minimizing the resource requirements
Proposed Algorithm
Transform Point Selection Goal : minimize the size of the census transform vector Challenge: Must quantify how much each point in the transform window contributes to overall correlation accuracy Test correlation accuracy: Define a sparse census transform consisting of a single point (| 𝑵 | = 1) Determine how consistently this point leads to correct correlation 13 × 13 correlation window (aggregation)
Tsukuba Venus Teddy Cones
Transform Point Selection Go Tsukuba Venus Average Bright: Higher correlation accuracy 25 × 25 neighborhood Teddy Cones
Proposed Sparse Census Transform Very good correlation accuracy can be achieved using very sparse transforms. 16-point 12-point 8-point 4-point 2-point 1-point
Experimental Results
Generalized Census Transform Goal : greater freedom in choosing the census transform design Definition : redrawing the transform as a graph 3 × 3 correlation (aggregation) 3 × 3 census
Generalized Census Transform As.. (1)transform neighborhoods become more and more sparse (2)fewer pixels are used in the correlation process selection of points to include in the transform becomes more critical 如..(1)轉換的鄰點變得越來越稀疏(2)較少的像素中的相關處理中使用 選點在變換變得更重要,以包括 Horizontal + Vertical 2-edge 2-point
Generalized Census Transform symmetric
Proposed Generalized Census Transform Benefits : Often require a smaller census transform window (memory) Increased robustness under varying conditions (noise) 16-edge 12-edge 8-edge 4-edge 2-edge 1-edge
Experimental Results
Experimental Results
Experimental Results
Left Image Ground Truth Full 7x7 census 12-edge 4-edge
Left Image Ground Truth Full 7x7 census 12-edge 4-edge
Left Image Full 7x7 census 12-edge 4-edge
Experimental Results 𝟖𝟖%↓ 𝟔𝟏%↓ LUTs (look-up tables) : the amount of logic required to implement the method FFs : the number of 1-bit registers (the amount of pipelining used) RAMs : the number of 18-kbit block memories Freq. : the maximum operating frequency reported by synthesis
Conclusion
Conclusion Proposed and analyzed in this paper: A range of sparse census transforms reduce hardware resource requirements attempting to maximize correlation accuracy. often better than or nearly as good as the full census Generalized census transforms increased robustness in the presence of image noise 提出並在此進行了分析: 稀疏CT的範圍 減少硬體資源要求 試圖最大限度的相關精度。 比Full CT 轉換來的好 一般化的CT 在圖像noise的情況下提高了穩定性,
Q&A