Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
162 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Robust Confidence Intervals in Stereo Matching using Possibility Theory (2404.06273v1)

Published 9 Apr 2024 in cs.CV

Abstract: We propose a method for estimating disparity confidence intervals in stereo matching problems. Confidence intervals provide complementary information to usual confidence measures. To the best of our knowledge, this is the first method creating disparity confidence intervals based on the cost volume. This method relies on possibility distributions to interpret the epistemic uncertainty of the cost volume. Our method has the benefit of having a white-box nature, differing in this respect from current state-of-the-art deep neural networks approaches. The accuracy and size of confidence intervals are validated using the Middlebury stereo datasets as well as a dataset of satellite images. This contribution is freely available on GitHub.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (38)
  1. Joint propagation of variability and imprecision in assessing the risk of groundwater contamination. Journal of Contaminant Hydrology, 93(1-4):72–84, 2007.
  2. l-, 2- and 3-dimensional modeling of water movement in the unsaturated soil matrix using a fuzzy approach. Advances in Water Resources, 1995.
  3. DeepSim-Nets: Deep Similarity Networks for Stereo Image Matching. In 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 2097–2105, Vancouver, BC, Canada, 2023. IEEE.
  4. Learning the Distribution of Errors in Stereo Matching for Joint Disparity and Uncertainty Estimation, 2023. arXiv:2304.00152 [cs].
  5. Ground Truth Generation and Disparity Estimation for Optical Satellite Imagery. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLIII-B2-2020:127–134, 2020.
  6. Random sets and fuzzy interval analysis. Fuzzy Sets and Systems, 42(1):87–101, 1991.
  7. When upper probabilities are possibility measures. Fuzzy Sets and Systems, 49(1):65–74, 1992.
  8. Improving Pairwise DSM With 3SGM: a Semantic Segmentation for SGM Using an Automatically Refined Neural Network. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, V-2-2022:167–175, 2022.
  9. MGM: A Significantly More Global Matching for Stereovision. In Procedings of the British Machine Vision Conference 2015, pages 90.1–90.12, Swansea, 2015. British Machine Vision Association.
  10. An automatic and modular stereo pipeline for pushbroom images. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, II-3:49–56, 2014.
  11. Are we ready for autonomous driving? The KITTI vision benchmark suite. In 2012 IEEE Conference on Computer Vision and Pattern Recognition, pages 3354–3361, Providence, RI, 2012. IEEE.
  12. Confidence Estimation for Superpixel-Based Stereo Matching. In 2015 International Conference on 3D Vision, pages 180–188, Lyon, France, 2015. IEEE.
  13. Group-Wise Correlation Stereo Network. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 3268–3277, Long Beach, CA, USA, 2019. IEEE.
  14. Real-time semi-global dense stereo solution with improved sub-pixel accuracy. In 2010 IEEE Intelligent Vehicles Symposium, pages 369–376, La Jolla, CA, USA, 2010. IEEE.
  15. Heiko Hirschmüller. Accurate and Efficient Stereo Processing by Semi-Global Matching and Mutual Information. In 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), pages 807–814, San Diego, CA, USA, 2005. IEEE.
  16. A Quantitative Evaluation of Confidence Measures for Stereo Vision. IEEE Transactions on Pattern Analysis and Machine Intelligence, 34(11):2121–2133, 2012.
  17. Geometry-Based Regularization For Dense Image Matching Via Uncertainty-Driven Depth Propagation. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, V-2-2020:151–159, 2020.
  18. A Survey on Deep Learning Techniques for Stereo-Based Depth Estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(4):1738–1764, 2022.
  19. Mapping snow depth in open alpine terrain from stereo satellite imagery. The Cryosphere, 10(4):1361–1380, 2016.
  20. Disparity Estimation Networks for Aerial and High-Resolution Satellite Images: A Review. Image Processing On Line, 12:501–526, 2022.
  21. CNN-Based Cost Volume Analysis as Confidence Measure for Dense Matching. In 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), pages 2070–2079, Seoul, Korea (South), 2019. IEEE.
  22. CO3D Mission Digital Surface Model Production Pipeline. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLIII-B2-2020, 2020.
  23. Mourad Oussalah. On the normalization of subnormal possibility distributions: New investigations. International Journal of General Systems, 31(3):277–301, 2002.
  24. Learning a General-Purpose Confidence Measure Based on O(1) Features and a Smarter Aggregation Strategy for Semi Global Matching. In 2016 Fourth International Conference on 3D Vision (3DV), pages 509–518, Stanford, CA, USA, 2016. IEEE.
  25. On the confidence of stereo matching in a deep-learning era: a quantitative evaluation. IEEE Transactions on Pattern Analysis and Machine Intelligence, pages 1–1, 2021.
  26. MicMac – a free, open-source solution for photogrammetry. Open Geospatial Data, Software and Standards, 2(1):14, 2017.
  27. Ambiguity Concept In Stereo Matching Pipeline. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLIII-B2-2021:383–390, 2021.
  28. Learning Conditional Random Fields for Stereo. In 2007 IEEE Conference on Computer Vision and Pattern Recognition, pages 1–8, Minneapolis, MN, USA, 2007. IEEE.
  29. High-accuracy stereo depth maps using structured light. In 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings., pages I–195–I–202, Madison, WI, USA, 2003. IEEE Comput. Soc.
  30. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. In Proceedings IEEE Workshop on Stereo and Multi-Baseline Vision (SMBV 2001), pages 131–140, Kauai, HI, USA, 2001. IEEE Comput. Soc.
  31. High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth. In Pattern Recognition, pages 31–42. Springer International Publishing, Cham, 2014. Series Title: Lecture Notes in Computer Science.
  32. An automated, open-source pipeline for mass production of digital elevation models (DEMs) from very-high-resolution commercial stereo satellite imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 116:101–117, 2016.
  33. Learning to Detect Ground Control Points for Improving the Accuracy of Stereo Matching. In 2014 IEEE Conference on Computer Vision and Pattern Recognition, pages 1621–1628, Columbus, OH, USA, 2014. IEEE.
  34. Leveraging Confident Points for Accurate Depth Refinement on Embedded Systems. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 158–167, Long Beach, CA, USA, 2019. IEEE.
  35. Peter Walley. Statistical Reasoning with Imprecise Probabilities. Springer US, Boston, MA, 1991.
  36. CARS: A Photogrammetry Pipeline Using Dask Graphs to Construct A Global 3D Model. In IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium, pages 453–456, Waikoloa, HI, USA, 2020. IEEE.
  37. Non-parametric local transforms for computing visual correspondence. In Computer Vision — ECCV ’94, pages 151–158. Springer Berlin Heidelberg, Berlin, Heidelberg, 1994. Series Title: Lecture Notes in Computer Science.
  38. Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches. Journal of Machine Learning Research 17, 2016. arXiv: 1510.05970.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets