Abstract
Stereo cameras are now commonly available on cars and mobile phones. However, the captured images may suffer from low image quality under noisy conditions, producing inaccurate disparity. In this paper, we aim at jointly restoring a clean image pair and estimating the corresponding disparity. To this end, we propose a new joint framework that iteratively optimizes these two different tasks in a multiscale fashion. First, structure information between the stereo pair is utilized to denoise the images using a non-local means strategy. Second, a new noise-tolerant cost function is proposed for noisy stereo matching. These two terms are integrated into a multiscale framework in which cross-scale information is leveraged to further improve both denoising and stereo matching. Extensive experiments on datasets captured from indoor, outdoor, and low-light conditions show that the proposed method achieves superior performance than the state-of-the-art image denoising and disparity estimation methods. While it outperforms multi-image denoising methods by about 2 dB on average, it achieves a 50% error reduction over radiometric-change-robust stereo matching on the challenging KITTI dataset.
Similar content being viewed by others
References
Alter, F., Matsushita, Y., & Tang, X. (2006). An intensity similarity measure in low-light conditions. In ECCV.
Birchfield, S., & Tomasi, C. (1998). A pixel dissimilarity measure that is insensitive to image sampling. IEEE TPAMI, 20(4), 401–406.
Buades, A., Coll, B., & Morel, J. (2005). A non-local algorithm for image denoising. In CVPR (pp. 60–65).
Buades, A., Coll, B., & Morel, J. M. (2005). A review of image denoising algorithms, with a new one. Multiscale Modeling and Simulation, 4(2), 490–530.
Buades, A., Coll, B., & Morel, J. M. (2008). Nonlocal image and movie denoising. IJCV, 76(2), 123–139.
Burger, H., Schuler, C., & Harmeling, S. (2012). Image denoising: Can plain neural networks compete with BM3D? In CVPR (pp. 2392–2399).
Chan, S., Zickler, T., & Lu, Y. (2014). Monte Carlo non-local means: Random sampling for large-scale image filtering. IEEE TIP, 23(8), 3711–3725.
Chen, F., Zhang, L., & Yu, H. (2015). External patch prior guided internal clustering for image denoising. In ICCV(pp. 603–611).
Dabov, K., Foi, A., Katkovnik, V., & Egiazarian, K. (2007). Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE TIP, 16(8), 2080–2095.
De-Maeztu, L., Villanueva, A., & Cabeza, R. (2011). Stereo matching using gradient similarity and locally adaptive support-weight. Pattern Recognition Letters, 32(13), 1643–1651.
Deledalle, C., Salmon, J., & Dalalyan, A. (2011). Image denoising with patch based PCA: Local versus global. In BMVC (pp. 425–455).
Dong, W., Li, G., Shi, G., Li, X., & Ma, Y. (2015). Low-rank tensor approximation with Laplacian scale mixture modeling for multiframe image denoising. In ICCV (pp. 442–449).
Felzenszwalb, P., & Huttenlocher, D. (2006). Efficient belief propagation for early vision. IJCV, 70(1), 41–54.
Fu, Y., Lam, A., Sato, I., & Sato, Y. (2016). Adaptive spatial-spectral dictionary learning for hyperspectral image restoration. IJCV, 122, 1–18.
Healey, G., & Kondepudy, R. (1994). Radiometric CCD camera calibration and noise estimation. IEEE TPAMI, 16(3), 267–276.
Heo, Y., Lee, K., & Lee, S. (2007). Simultaneous depth reconstruction and restoration of noisy stereo images using non-local pixel distribution. In CVPR (pp. 1–8).
Heo, Y., Lee, K., & Lee, S. (2011). Robust stereo matching using adaptive normalized cross-correlation. IEEE TPAMI, 33(4), 807–822.
Hirschmüller, H. (2008). Stereo processing by semiglobal matching and mutual information. IEEE TPAMI, 30(2), 328–341.
Hirschmüller, H., Innocent, P., & Garibaldi, J. (2002). Real-time correlation-based stereo vision with reduced border errors. IJCV, 47(1), 229–246.
Hirschmüller, H., & Scharstein, D. (2009). Evaluation of stereo matching costs on images with radiometric differences. IEEE TPAMI, 31(9), 1582–1599.
Honda, H., Timofte, R., & Gool, L.V. (2015). Make my day—high-fidelity color denoising with near-infrared. In CVPR workshops (pp. 82–90).
Joshi, N., & Cohen, M. (2010). Seeing Mt. Rainier: Lucky imaging for multi-image denoising, sharpening, and haze removal. In ICCP (pp. 1–8).
Jung, I., Sim, J., Kim, C., & Lee, S. (2013). Robust stereo matching under radiometric variations based on cumulative distributions of gradients. In ICIP (pp. 2082–2085).
Kim, Y., Koo, J., & Lee, S. (2016). Adaptive descriptor-based robust stereo matching under radiometric changes. Pattern Recognition Letters, 78, 41–47.
Levin, A., & Nadler, B. (2011). Natural image denoising: Optimality and inherent bounds. In CVPR (pp. 2833–2840).
Levin, A., Nadler, B., Durand, F., & Freeman, W. (2012). Patch complexity, finite pixel correlations and optimal denoising. In ECCV (pp. 73–86).
Liu, C., & Freeman, W. (2010). A high-quality video denoising algorithm based on reliable motion estimation. In ECCV (pp. 706–719).
Lu, X., Lin, Z., Jin, H., Yang, J., & Wang, J. (2015). Image-specific prior adaptation for denoising. IEEE TIP, 24(12), 5469–5478.
Luo, E., Chan, S., & Nguyen, T. (2015). Adaptive image denoising by targeted databases. IEEE TIP, 24(7), 2167–2181.
Luo, E., Chan, S., Pan, S., & Nguyen, T. (2013). Adaptive non-local means for multiview image denoising: Searching for the right patches via a statistical approach. In ICIP (pp. 543–547).
Maggioni, M., Katkovnik, V., Egiazarian, K., & Foi, A. (2013). Nonlocal transform-domain filter for volumetric data denoising and reconstruction. IEEE TIP, 22(1), 119–133.
Menz, M., & Freeman, R. (2003). Stereoscopic depth processing in the visual cortex: a coarse-to-fine mechanism. Nature Neuroscience, 6(1), 59–65.
Menze, M., & Geiger, A. (2015). Object scene flow for autonomous vehicles. In CVPR (pp. 3061–3070).
Mosseri, I., Zontak, M., & Irani, M. (2013). Combining the power of internal and external denoising. In ICIP (pp. 1–9).
Muresan, D., & Parks, T. (2003). Adaptive principal components and image denoising. In ICIP (pp. 101–104).
Nir, T., Kimmel, R., & Bruckstein, A. (2005). Variational approach for joint optic-flow computation and video restoration. Technical report CIS200503, Technion, Israel Institute of Technology.
Park, B., Lee, K., & Lee, S. (2006). A new similarity measure for random signatures: Perceptually modified Hausdorff distance. In ACIVS (pp. 990–1001).
Romeny, B. M. T. H., & Florack, L. (1993). A multiscale geometric model of human vision. In W. R. Hendee & P. N. T. Wells (Eds.), Perception of visual information, Chap. 4 (pp. 73–114).
Scharstein, D., & Szeliski, R. (2002). A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. IJCV, 47(1), 7–42.
Scharstein, D., & Szeliski, R. (2007). Middlebury stereo dataset. http://vision.middlebury.edu/stereo/data/.
Shao, L., Yan, R., & Li, X. (2014). From heuristic optimization to dictionary learning: A review and comprehensive comparison of image denoising algorithms. IEEE Transsction on Cybernetics, 44(7), 1001–1013.
Shen, X., Yan, Q., Xu, L., Ma, L., & Jia, J. (2015). Multispectral joint image restoration via optimizing a scale map. IEEE TPAMI, 37(12), 2518–2530.
Tan, X., Sun, C., Wang, D., Guo, Y., & Pham, T. (2014). Soft cost aggregation with multi-resolution fusion. In ECCV (pp. 17–32).
Vemulapalli, R., Tuzel, O., & Liu, M. (2015). Deep Gaussian conditional random field network: A model-based deep network for discriminative denoising. arXiv:1511.04067.
Vu, D., Chidester, B., Yang, H., Do, M., & Lu, J. (2014). Efficient hybrid tree-based stereo matching with applications to postcapture image refocusing. IEEE TIP, 23(8), 3428–3442.
Xu, J., Yang, Q., Tang, J., & Feng, Z. (2016). Linear time illumination invariant stereo matching. IJCV, 119, 179–193.
Xu, J., Zhang, L., Zuo, W., Zhang, D., & Feng, X. (2015). Patch group based nonlocal self-similarity prior learning for image denoising. In ICCV (pp. 244–252).
Xu, L., & Jia, J. (2010). Two-phase kernel estimation for robust motion deblurring. In ECCV (pp. 157–170).
Yue, H., Sun, X., Yang, J., & Wu, F. (2014). Cid: Combined image denoising in spatial and frequency domains using web images. In CVPR (pp. 2933–2940).
Yue, H., Sun, X., Yang, J., & Wu, F. (2015). Image denoising by exploring external and internal correlations. IEEE TIP, 24(6), 1967–1982.
Zabih, R., & Woodfill, J. (1994). Non-parametric local transforms for computing visual correspondence. In ECCV (pp. 151–158).
Zbontar, J., & LeCun, Y. (2016). Stereo matching by training a convolutional neural network to compare image patches. Journal of Machine Learning Research, 17(1), 2287–2318.
Zhang, K., Fang, Y., Min, D., Sun, L., Yang, S., Yan, S., et al. (2014). Cross-scale cost aggregation for stereo matching. In CVPR (pp. 1590–1597).
Zhang, L., Dong, W., Zhang, D., & Shi, G. (2010). Two-stage image denoising by principal component analysis with local pixel grouping. Pattern Recognition, 43(4), 1531–1549.
Zhang, L., Vaddadi, S., Jin, H., & Nayar, S. (2009). Multiple view image denoising. In CVPR (pp. 1542–1549).
Zontak, M., Mosseri, I., & Irani, M. (2013). Separating signal from noise using patch recurrence across scales. In CVPR (pp. 1195–1202).
Zoran, D., & Weiss, Y. (2011) From learning models of natural image patches to whole image restoration. In ICCV (pp. 479–486).
Acknowledgements
We would thank the anonymous reviewers for their constructive suggestions and insightful comments. This work is partially supported by the Hong Kong PhD Fellowship Scheme (HKPFS) from the RGC of Hong Kong, a SRG grant from City University of Hong Kong (No. 7004416), and a GRF grant from the RGC of Hong Kong (PolyU 52124/15E).
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Scharstein.
Rights and permissions
About this article
Cite this article
Jiao, J., Yang, Q., He, S. et al. Joint Image Denoising and Disparity Estimation via Stereo Structure PCA and Noise-Tolerant Cost. Int J Comput Vis 124, 204–222 (2017). https://doi.org/10.1007/s11263-017-1015-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11263-017-1015-9