45
Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

Embed Size (px)

Citation preview

Page 1: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

1

Window-based Approach For Fast Stereo Correspondence

Raj Kumar Gupta, Siu-Yeung Cho

IET Computer Vision,  2013

Page 2: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

2

Outline•Introduction•Related Work•Proposed Method•Experimental Results•Conclusion

Page 3: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

3

Introduction•Using two correlation windows to improve the performance of the algorithm •3*3 and 9*9

•Real-time suitability •more than 10 frame/s on CPU in case of 320 × 240-sized image pair with disparity value 16

Page 4: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

4

Related Work•Local methods are usually base on correlation.•Area-based (NCC, SAD, SSD)•Feature-based: rely on feature extraction and match local cues (BF, GF)

•Bigger window size, more information, more blurred.

Page 5: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

5

Outline•Introduction•Related Work•Proposed Method•Experimental Results•Conclusion

Page 6: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

6

Flow Chart

Page 7: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

7

Flow Chart

Page 8: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

8

Initial Matching•Matching cost computation: SAD

Left Right

d

Page 9: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

9

Problem in disparity selection

a. Determine disparity easily for unique minimum valueb. Ambiguous disparity in case of multiple minimac. Matching cost calculated at point (205, 230) of Tsukuba image

Page 10: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

10

Initial Matching: large correlation window

•Matching cost computation: SAD + penalty

•Penalty term

• Disparity computation

Page 11: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

11

Problem in disparity selection

Page 12: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

12

Initial Matching: small correlation window

•Only those disparity values that are carried by neighbouring pixels.•Matching cost computation without penalty

•N: the disparity values of the neighbouring pixels.•Avoid local minima and speed up

Page 13: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

13

Flow Chart

Page 14: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

14

Unreliable pixel detection

• left–right cross-checking

Page 15: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

15

Disparity Interpolation•Search for pixels with reliable disparity value in its eight neighbouring pixels.•Compute similarity of unreliable pixel and its reliable neighbor.

Page 16: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

16

Flow Chart

Page 17: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

17

Disparity Refinement

Page 18: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

18

Outline•Introduction•Related Work•Proposed Method•Experimental Results•Conclusion

Page 19: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

19

Experimental Results•Computation time of the proposed algorithm for different window sizes on Tsukuba image.(image size 384 × 288 with 16 disparity labels)

Page 20: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

20

• Percentage error in non-occluded (nocc), whole image (all) and near depth discontinuities (disc) for different window sizes for all four images (Tsukuba, Venus, Teddy and Cones)

Page 21: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

21

Experimental Results

• a. Without using small correlation window Ws and the disparity refinement step• b. Without using the disparity refinement step• c. Without using small correlation window Ws• d. With all four steps on Tsukuba image

Page 22: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

22

Experimental Results

Page 23: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

23

Page 24: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

24

Page 25: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

25

Experimental Results•Comparison the performance of the proposed algorithm with other correlation-based algorithms.

Page 26: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

26

Experimental Results

Page 27: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

27

Reference• [24] Gupta, R., Cho, S.-Y.: ‘Real-time stereo matching using adaptive binary window’

(3D Data Processing, Visualization and Transmission, 2010)• [25] Zhang, K., Lu, J., Lafruit, G., Lauwereins, R., Gool, L.V.: ‘Real-time accurate

stereo with bitwise fast voting on Cuda’. Int. Conf. Computer Vision Workshops, 2009, pp. 540–547• [26] Humenberger, M., Zinner, C., Weber, M., Kubinger, W., Vincze, M.:‘A fast

stereo matching algorithm suitable for embedded real-time systems’, Comput. Vis. Image Underst., 2010, 114, (11),pp. 1180–1202• [27] Gong, M., Yang, Y.: ‘Near real-time reliable stereo matching using

programmable graphics hardware’. IEEE Conf. Computer Vision and Pattern Recognition, 2005, pp. 924–931• [28] Richardt, C., Orr, D., Davies, I., Criminisi, A., Dodgson, N.: ‘Real-time

spatiotemporal stereo matching using the dual-cross-bilateral grid’. European Conf. Computer Vision, 2010, vol. 6313, pp. 510–523• [29] Ambrosch, K., Kubinger, W.: ‘Accurate hardware-based stereo vision’,Comput.

Vis. Image Underst., 2010, 114, (11), pp. 1303–1316

Page 28: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

28

Page 29: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

29

Page 30: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

30

Page 31: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

31

Page 32: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

32

Page 33: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

33

Page 34: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

34

Conclusion•A new correlation-based stereo-matching approach.• Large window improves at non-textured image regions• Small window improves at depth discontinuities

•The CPU implementation computes at a speed of more than 10 frame/s.•Easily implemented on GPU. •The proposed method can be used in real-time applications to reconstruct the 3D structures with great accuracy at object boundaries.

Page 35: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

35

Codebook based Stereo Matching for Natural User

Interface

Sung-il Kang and Hyunki Hong

2013 IEEE International Conference on Consumer Electronics (ICCE)

Page 36: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

36

Outline•Introduction•Proposed Method•Experimental Results•Conclusion

Page 37: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

37

Introduction

•Interactive user interface has been one of the major topics in consumer electronics.•Gesture based user interface• Interactive smart TV, Nintendo Wii, Sony PlayStation3 Move, and Microsoft Kinect.

•Propose a stereo system implemented on GPGPU for real-time performance.•Employ codebook to solve occlusion.

Page 38: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

38

Flow chart

Page 39: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

39

Proposed Method•Pre-processing• Laplace od Gaussian (LoG) filter for alleviating the lighting effects.

•Cost initialization• AD+Census[6]

[6] X. Mei, X. Sun, M. Zhou, H. Wang, and X. Zhang, “On building an accurate stereo matchng system on graphics hardware,” Proc. of GPUCV, pp. 467-474, 2011. http://www.camdemy.com/media/4724

Page 40: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

40

Proposed Method•Cost aggregation[6]

• Cross-based aggregation• Color similarity and the length constraint

•Refinement[6,7]

• Left-right consistency check• Iterative region voting• Sub-pixel enhancement

d

[7] Q. Yang, C. Engels, R. Yang, H. Stewenius, and D. Nister, “Stereo matching with color-weighted correlation, hierarchical belief propagation and occlusion handling,” IEEE Transactions on PAMI, 2009.

Page 41: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

41

Proposed Method

Occlusion? Find codeword

Updatecodeword

Codeword?

Yes

No

Yes

No

Add a newcodeword

Page 42: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

42

Experimental Results• Device: Intel Quad 2.66GHz with Nvidia GTX460. • Stereo images are captured by a Bumblebee 3 from Point Grey Inc. • Time: 80~110ms/frame• Stereo matching is implemented on GPU.• The codebook generation and its evaluation is on CPU.

Page 43: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

43

Experimental Results

Page 44: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

44

Experimental Results

[8] K. J. Yoon and I. S. Kweon, “Adaptive support-weight approach for correspondence search,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 28, pp. 650-656, 2005.[9] C. Richardt, D Orr, I Davies, and A Criminisi, “Real-time spatiotemporal stereo matching using the dual-cross-bilateral grid,” Proc. of ECCV, 2010.

Page 45: Window-based Approach For Fast Stereo Correspondence Raj Kumar Gupta, Siu-Yeung Cho IET Computer Vision, 2013 1

45

Conclusion•Propose a stereo system implemented on GPGPU for real-time performance.•Good performance at static background Only.