Unsupervised Learning of Depth and Ego-Motion from Monocular Video Using 3D Geometric Constraints
Open accessAbstract
We present a novel approach for unsupervised learning of depth and ego-motion from monocular video. Unsupervised learning removes the need for separate supervisory signals (depth or ego-motion ground truth, or multi-view video). Prior work in unsupervised depth learning uses pixel-wise or gradient-based losses, which only consider pixels in small local neighborhoods. Our main contribution is to explicitly consider the inferred 3D geometry of the whole scene, and enforce consistency of the estimated 3D point clouds and ego-motion across consecutive frames. This is a challenging task and is solved by a novel (approximate) backpropagation algorithm for aligning 3D structures. We combine this novel 3D-based loss with 2D losses based on photometric quality of frame reconstructions using estimated depth and ego-motion from adjacent frames. We also incorporate validity masks to avoid penalizing areas in which no useful information exists. We test our algorithm on the KITTI dataset and on a video dataset captured on an uncalibrated mobile phone camera. Our proposed approach consistently improves depth estimates on both datasets, and outperforms the state-of-the-art for both depth and ego-motion. Because we only require a simple video, learning depth and ego-motion on large and varied datasets becomes possible. We demonstrate this by training on the low quality uncalibrated video dataset and evaluating on KITTI, ranking among top performing prior methods which are trained on KITTI itself.
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.
Integrity signals
Registry
Cites
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.
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?
ImageNet pretraining improves accuracy for the compact model compared to training from scratch.
“Similar to [30,16], we start with weights pretrained on Ima-geNet [54], and show that this improves accuracy for our compact model compared to training from scratch (Table 2).”
From Digging Into Self-Supervised Monocular Depth Estimation (2019) · cites this paper as [54]
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
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404 · see citing works
Image quality assessment: from error visibility to structural similarity
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004) · IEEE Transactions on Image Processing · doi:10.1109/tip.2003.819861 · Cited by 57,892 · see citing works
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834 · see citing works
A method for registration of 3-D shapes
Besl, P. J., & McKay, N. D. (1992) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.121791 · Cited by 18,363 · see citing works
TensorFlow: A system for large-scale machine learningPreprint
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., Isard, M., et al. (2016) · arXiv (Cornell University) · doi:10.5555/3026877.3026899 · Cited by 16,369 · see citing works
Self-citationAre we ready for autonomous driving? The KITTI vision benchmark suite
Geiger, A., Lenz, P., & Urtasun, R. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6248074 · Cited by 14,897 · see citing works
The Cityscapes Dataset for Semantic Urban Scene Understanding
Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., & Schiele, B. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.350 · Cited by 12,433 · see citing works
Vision meets robotics: The KITTI dataset
Geiger, A., Lenz, P., Stiller, C., & Urtasun, R. (2013) · The International Journal of Robotics Research · doi:10.1177/0278364913491297 · Cited by 9,913 · see citing works
A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms
Scharstein, D., & Szeliski, R. (2002) · International Journal of Computer Vision · doi:10.1023/a:1014573219977 · Cited by 6,740 · see citing works
3D is here: Point Cloud Library (PCL)
Rusu, R. B., & Cousins, S. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980567 · Cited by 4,980 · see citing works
Efficient variants of the ICP algorithm
Rusinkiewicz, S., & Levoy, M. (2002) · Proceedings Third International Conference on 3-D Digital Imaging and Modeling · doi:10.1109/im.2001.924423 · Cited by 3,945 · see citing works
Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
Eigen, D., Puhrsch, C., & Fergus, R. (2014) · arXiv · doi:10.48550/arxiv.1406.2283 · Cited by 3,442 · see citing works
Unsupervised Monocular Depth Estimation with Left-Right Consistency
Godard, C., Aodha, O. M., & Brostow, G. J. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.699 · Cited by 3,270 · see citing works
Unsupervised Learning of Depth and Ego-Motion from Video
Zhou, T., Brown, M., Snavely, N., & Lowe, D. G. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.700 · Cited by 2,910 · see citing works
Loss Functions for Image Restoration With Neural Networks
Zhao, H., Gallo, O., Frosio, I., & Kautz, J. (2016) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2016.2644865 · Cited by 2,659 · see citing works
Mayer, N., Ilg, E., Hausser, P., Fischer, P., Cremers, D., Dosovitskiy, A., & Brox, T. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.438 · Cited by 2,488 · see citing works
Deeper Depth Prediction with Fully Convolutional Residual Networks
Laina, I., Rupprecht, C., Belagiannis, V., Tombari, F., & Navab, N. (2016) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2016.32 · Cited by 2,061 · see citing works
Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue
Garg, R., B.G., V. K., Carneiro, G., & Reid, I. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46484-8_45 · Cited by 1,632 · see citing works
Object modeling by registration of multiple range images
Chen, Y., & Medioni, G. (2002) · Proceedings. 1991 IEEE International Conference on Robotics and Automation · doi:10.1109/robot.1991.132043 · Cited by 1,409 · see citing works
The Cityscapes Dataset for Semantic Urban Scene Understanding
Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., & Schiele, B. (2016) · arXiv · doi:10.48550/arxiv.1604.01685 · Cited by 1,174 · see citing works
DeMoN: Depth and Motion Network for Learning Monocular Stereo
Ummenhofer, B., Zhou, H., Uhrig, J., Mayer, N., Ilg, E., Dosovitskiy, A., & Brox, T. (2016) · Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2017.596 · Cited by 777 · see citing works
Li, B., Shen, C., Dai, Y., van den Hengel, A., & He, M. (2015) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2015.7298715 · Cited by 621 · see citing works
Deep Stereo: Learning to Predict New Views from the World's Imagery
Flynn, J., Neulander, I., Philbin, J. F., & Snavely, N. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.595 · Cited by 599 · see citing works
Pulling Things out of Perspective
Ladický, Ľ., Shi, J., & Pollefeys, M. (2014) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2014.19 · Cited by 463 · see citing works
(2016) · doi:10.1007/978-3-319-46493-0_51 · Cited by 411 · see citing works
Learning a Multi-View Stereo MachinePreprint
Kar, A., Häne, C., & Malik, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1708.05375 · Cited by 331 · see citing works
SfM-Net: Learning of Structure and Motion from VideoPreprint
Vijayanarasimhan, S., Ricco, S., Schmid, C., Sukthankar, R., & Fragkiadaki, K. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1704.07804 · Cited by 327 · see citing works
Learning a Multi-View Stereo Machine
Kar, A., Häne, C., & Malik, J. (2017) · neural information processing systems · Cited by 248 · see citing works
Loss Functions for Neural Networks for Image ProcessingPreprint
Zhao, H., Gallo, O., Frosio, I., & Kautz, J. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1511.08861 · Cited by 236 · see citing works
Dense Monocular Depth Estimation in Complex Dynamic Scenes
Ranftl, R., Vineet, V., Chen, Q., & Koltun, V. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.440 · Cited by 174 · see citing works
PM-Huber: PatchMatch with Huber Regularization for Stereo Matching
Heise, P., Klose, S., Jensen, B., & Knoll, A. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.293 · Cited by 170 · see citing works
Is L2 a Good Loss Function for Neural Networks for Image ProcessingPreprint
Zhao, H., Gallo, O., Frosio, I., & Kautz, J. (2015) · arXiv (Cornell University) · Cited by 36 · see citing works
Flagged first, then by citations
Research works printed but not linked: 5
- Y. Cao, Z. Wu, , and C. Shen. Estimating depth from monoc- ular images as classification using deep fully convolutional residual networks. CoRR:1605.02305, 2016. 2
- J. Flynn, I. Neulander, J. Philbin, and N. Snavely. Deep- stereo: Learning to predict new views from the worlds im- agery. In CVPR, 2016. 2
- M. Jaderberg, K. Simonyan, A. Zisserman, et al. Spatial transformer networks. In Advances in Neural Information Processing Systems, pages 2017-2025, 2015. 3
- J. Xie, R. Girshick, and A. Farhadi. Deep3d: Fully au- tomatic 2d-to-3d video conversion with deep convolutional neural networks. ECCV, 2016. 2
- J. Zbontar and Y. LeCun. Stereo matching by training a con- volutional neural network to compare image patches. JMLR, 2016. 2
Cited by
Sorted by citations
Digging Into Self-Supervised Monocular Depth Estimation
Godard, C., Aodha, O. M., Firman, M., & Brostow, G. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00393 · Cited by 2,561 · see citing works
Zhang, J., & Tao, D. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3039359 · Cited by 774 · see citing works
3D Packing for Self-Supervised Monocular Depth Estimation
Guizilini, V., Ambrus, R., Pillai, S., Raventos, A., & Gaidon, A. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00256 · Cited by 741 · see citing works
D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometry
Yang, N., von Stumberg, L., Wang, R., & Cremers, D. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00136 · Cited by 438 · see citing works
Depth From Videos in the Wild: Unsupervised Monocular Depth Learning From Unknown Cameras
Gordon, A., Li, H., Jonschkowski, R., & Angelova, A. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00907 · Cited by 389 · see citing works
Self-citationConsistent video depth estimation
Luo, X., Huang, J.-B., Szeliski, R., Matzen, K., & Kopf, J. (2020) · ACM Transactions on Graphics · doi:10.1145/3386569.3392377 · Cited by 322 · see citing works
MonoGRNet: A Geometric Reasoning Network for Monocular 3D Object Localization
Qin, Z., Wang, J., & Lu, Y. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33018851 · Cited by 290 · see citing works
Self-Supervised Monocular Depth Hints
Watson, J., Firman, M., Brostow, G., & Turmukhambetov, D. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00225 · Cited by 275 · see citing works
Unsupervised Adversarial Depth Estimation Using Cycled Generative Networks
Pilzer, A., Xu, D., Puscas, M., Ricci, E., & Sebe, N. (2018) · 2018 International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2018.00073 · Cited by 192 · see citing works
Wang, K., Ma, S., Chen, J., Ren, F., & Lu, J. (2022) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2020.3038898 · Cited by 114 · see citing works
Web Stereo Video Supervision for Depth Prediction from Dynamic Scenes
Wang, C., Lucey, S., Perazzi, F., & Wang, O. (2019) · 2019 International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2019.00046 · Cited by 103 · see citing works
Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery
Shamwell, E. J., Lindgren, K., Leung, S., & Nothwang, W. D. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2909895 · Cited by 86 · see citing works
- Authors
- Mahjourian, R., Wicke, M., & Angelova, A.
- Affiliations
- The University of Texas at Austin (US)
- Venue
- 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
- Publisher
- IEEE
- Published
- 2018-06
- Pages
- 5667-5675
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2785512290
- Cited by
- 775
- References
- 32
- Citations by year
- 2026: 18 · 2025: 40 · 2024: 50 · 2023: 85 · 2022: 123 · 2021: 151 · 2020: 147 · 2019: 137 · 2018: 24
- FWCI
- 34.67
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Physical Sciences › Engineering › Aerospace Engineering › Robotics and Sensor-Based Localization
- Physical Sciences › Earth and Planetary Sciences › Geology › 3D Surveying and Cultural Heritage
- Keywords
- unsupervised depth estimation, camera motion estimation, monocular video, KITTI dataset
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
