CoChat
Sign Up
Cited by 924Advanced Neural Network ApplicationsAdvanced Vision and Imaging3D Shape Modeling and Analysis

Sparsity Invariant CNNs

Open access
Uhrig, Jonas, Schneider, Nick, Schneider, Lukas, Franke, Uwe, Brox, Thomas, Geiger, Andreas·2017·2017 International Conference on 3D Vision (3DV)
DOI 10.1109/3dv.2017.00012
Open this paper in CoChat — free accountOpen PDF

Abstract

In this paper, we consider convolutional neural networks operating on sparse inputs with an application to depth completion from sparse laser scan data. First, we show that traditional convolutional networks perform poorly when applied to sparse data even when the location of missing data is provided to the network. To overcome this problem, we propose a simple yet effective sparse convolution layer which explicitly considers the location of missing data during the convolution operation. We demonstrate the benefits of the proposed network architecture in synthetic and real experiments with respect to various baseline approaches. Compared to dense baselines, the proposed sparse convolution network generalizes well to novel datasets and is invariant to the level of sparsity in the data. For our evaluation, we derive a novel dataset from the KITTI benchmark, comprising over 94k depth annotated RGB images. Our dataset allows for training and evaluating depth completion and depth prediction techniques in challenging real-world settings and is available online at: www.cvlibs.net/datasets/kitti.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/3dv.2017.00012

Cites

65 of 65 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Using the online refinement method of Casser et al. yields a significant improvement in depth estimation.

    “Using the online refinement method from [42], we obtain a significant improvement.”

    From OmniDet: Surround View Cameras Based Multi-Task Visual Perception Network for Autonomous Driving (2021) · cites this paper as [42]

    The Conclusion (p. 8) attributes improved performance to sparse convolution modules, while Boosting Single-Image Depth Prediction (pp. 14–15) discusses training with different depth maps. Neither passage addresses Casser et al.’s online refinement method or an improvement from using it; searches for “Casser” and “online refinement” found no such passage in our copy. (LLM, conf=0.76)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 17

  1. [2][2] D. Chan, H. Buisman, C. Theobalt, and S. Thrun. A NoiseAware Filter for Real-Time Depth Upsampling. In ECCV Workshops , 2008. 3 -
  2. [4][4] T. S. Cohen and M. Welling. Group equivariant convolutional networks. In Proc. of the International Conf. on Machine learning (ICML) , 2016. 2 -
  3. [6][6] J. Diebel and S. Thrun. An application of markov random fields to range sensing. In Advances in Neural Information Processing Systems (NIPS) , 2005. 3 -
  4. [13][13] M. Figurnov, A. Ibraimova, D. P. Vetrov, and P. Kohli. Perforatedcnns: Acceleration through elimination of redundant convolutions. In Advances in Neural Information Processing Systems (NIPS) , 2016. 2 -
  5. [21][21] S. Han, J. Pool, J. Tran, and W. J. Dally. Learning both weights and connections for efficient neural network. In Advances in Neural Information Processing Systems (NIPS) , 2015. 2 -
  6. [25][25] J. F. Henriques and A. Vedaldi. Warped convolutions: Efficient invariance to spatial transformations. arXiv.org , 1609.04382, 2016. 2 -
  7. [29][29] M. Jaderberg, K. Simonyan, A. Zisserman, and K. Kavukcuoglu. Spatial transformer networks. In Advances in Neural Information Processing Systems (NIPS) , 2015. 2 -
  8. [31][31] V. Jampani, M. Kiefel, and P. V. Gehler. Learning sparse high dimensional filters: Image filtering, dense CRFs and bilateral neural networks. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , pages 4452-4461, 2016. 7 -
  9. [32][32] J. Kim, J. K. Lee, and K. M. Lee. Accurate image superresolution using very deep convolutional networks. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , 2016. 3 -
  10. [35][35] A. Krizhevsky, I. Sutskever, and G. E. Hinton. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems (NIPS) , 2012. 2, 14 -
  11. [37][37] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. Gradientbased learning applied to document recognition. Proceedings of the IEEE , 86(11):2278-2324, 1998. 13 -
  12. [40][40] B. Liu, M. Wang, H. Foroosh, M. Tappen, and M. Pensky. Sparse convolutional neural networks. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , June 2015. 2 -
  13. [41][41] F. Liu, C. Shen, G. Lin, and I. Reid. Learning depth from single monocular images using deep convolutional neural fields. IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI) , 38(10):2024-2039, 2016. 14, 15 -
  14. [44][44] J. Long, E. Shelhamer, and T. Darrell. Fully convolutional networks for semantic segmentation. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , 2015. 12, 14 -
  15. [45][45] O. Mac Aodha, N. D. Campbell, A. Nair, and G. J. Brostow. Patch based synthesis for single depth image superresolution. In Proc. of the European Conf. on Computer Vision (ECCV) , pages 71-84. Springer, 2012. 3 -
  16. [67][67] J. Yang, J. Wright, T. S. Huang, and Y. Ma. Image superresolution via sparse representation. IEEE transactions on image processing , 19(11):2861-2873, 2010. 3 -
  17. [70][70] S. Zweig and L. Wolf. Interponet, A brain inspired neural network for optical flow dense interpolation. arXiv.org , 1611.09803, 2016. 2

Cited by

Sorted by citations

Related

Authors
Uhrig, J., Schneider, N., Schneider, L., Franke, U., Brox, T., & Geiger, A.
Affiliations
  • University of Freiburg (DE)
  • Karlsruhe Institute of Technology (DE)
  • ETH Zurich (CH)
  • Daimler (Germany) (DE)
  • Max Planck Society (DE)
ORCID
Venue
2017 International Conference on 3D Vision (3DV)
Publisher
IEEE
Published
2017-10
Pages
11-20
DOI
10.1109/3dv.2017.00012
OpenAlex
W2750215633
Cited by
924
References
65
Citations by year
2026: 52 · 2025: 90 · 2024: 121 · 2023: 138 · 2022: 134 · 2021: 143 · 2020: 114 · 2019: 99 · 2018: 33
FWCI
19.68
Type
conference paper
Language
en
Open access
Locations
  • International Conference on 3D Vision (3DV) (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
  • Physical Sciences › Engineering › Computational Mechanics › 3D Shape Modeling and Analysis
Keywords
sparse convolution, sparse input, depth completion, depth estimation, convolutional neural networks, KITTI dataset, RGB images

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review