Enhancing self-supervised monocular depth estimation with traditional visual odometry

Alessandro Pieropan; Belen Luque; Lorenzo Andraghetti; Matteo Poggi; Panteleimon Myriokefalitakis; Pier Luigi Dovesi; Stefano Mattoccia

REVIEW 3 major objections 3 minor 52 references

Enhancing self-supervised monocular depth estimation with traditional visual odometry

T0 review · 3 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Sparse 3D points from visual odometry, densified by a sparsity-invariant autoencoder, improve self-supervised monocular depth estimation on KITTI.

desk verdict Useful, well-ablated idea—fusing a sparse VO prior into self-supervised depth networks—but the headline SOTA comparison likely mixes test splits, so treat the central claim with caution. read the letter →

arxiv 1908.03127 v2 pith:JADJ44Z4 submitted 2019-08-08 cs.CV

Lorenzo Andraghetti , Panteleimon Myriokefalitakis , Pier Luigi Dovesi , Belen Luque , Matteo Poggi , Alessandro Pieropan , Stefano Mattoccia This is my paper · ORCID

classification cs.CV

keywords self-supervisedmonoculardepthestimationvisualodometrysparsity-invariantautoencodergeometricpriorcompletionstereoself-supervisionKITTIdatasetembeddeddeployment

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that a single-image depth estimator trained without ground truth can be made more accurate by feeding it, at inference time, sparse 3D points produced by a traditional visual odometry pipeline. The authors build a sparsity-invariant autoencoder that densifies those points, then add them as an extra input to two existing self-supervised stereo-trained networks, Monodepth and PyD-Net. On the KITTI odometry test split they report Abs Rel 0.091 against 0.108 for Monodepth, with similar gains for the compact PyD-Net, at a modest runtime cost. This matters because it points to a cheap way to improve monocular depth in autonomous navigation without requiring LiDAR or stereo at deployment.

What carries the argument

The machine is a sparsity-invariant autoencoder: five layers of sparse convolutions with 16 filters and decreasing kernel sizes (9x9, 5x5, 3x3, 3x3, then a pixel-wise layer) that densify very sparse VO disparity maps containing about 0.06% of pixels. This densified map is concatenated with the RGB image and fed both to the depth estimator and to a shared residual skip module that produces a correction added to the network's disparity output. Two losses anchor the prediction to the VO points: an inner loss on the autoencoder output and an outer loss on the final depth.

What would settle it

Run the same VOMonodepth pipeline with an open-source monocular VO with scale recovery on the same 8691-frame KITTI odometry test split and compare Abs Rel: if the gain over Monodepth's 0.108 disappears or turns negative, the reported advantage is tied to the private VO system rather than to the fusion architecture.

Watch

Extended reading notes

Core claim

The central claim is that a geometrical prior from a monocular VO algorithm, converted to sparse disparity maps, can be fused into a self-supervised depth network by a sparse-convolution autoencoder plus a residual skip module, and that this consistently improves accuracy over the base network. The design keeps the pipeline monocular at deployment: stereo pairs are used only during training, for self-supervision and to provide scale-aware VO cues, while at test time only the current image and VO points are needed. The paper reports that this strategy outperforms published self-supervised depth methods on the KITTI Eigen-split odometry subset, including MonoResMatch which uses additional SGM supervision, and that the compact variant still runs at 18 FPS on a Jetson TX2 without post-processing.

Load-bearing premise

The approach depends on a monocular visual odometry system that returns sparse 3D points with reliable scale during deployment; in the paper that system is a proprietary pipeline, not described or released.

Editorial extensions

If this is right

Self-supervised depth networks can be improved at inference time with purely geometrical cues from a single moving camera, without ground truth or stereo at deployment.
The same VO-prior injection works on both a ResNet-based network with about 58 million parameters and a compact network with fewer than 2 million parameters, so the gain is not tied to one architecture.
The runtime penalty is moderate: VOPyD-Net drops from 24.22 to 18.48 FPS on a Jetson TX2 without post-processing, remaining real-time.
Using an accurate scale-recovered monocular VO is enough to keep the pipeline monocular at test time, opening use in autonomous driving and embedded platforms.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the proprietary monocular VO used in the experiments were replaced by an open-source alternative with comparable scale recovery, the reported gains could shrink or vanish; the paper does not quantify sensitivity to VO noise.
The same fusion idea could be applied to other sparse geometric cues, such as feature tracks from SLAM or sparse LiDAR, reusing the sparsity-invariant autoencoder for any low-density prior.
Because the qualitative results show the strongest improvements on thin structures and vegetation edges, a testable extension is to measure gains separately on poles, trees, and distant objects.
The method implicitly assumes that VO points are correctly associated with the current frame; failure cases such as dynamic objects or degenerate motion, like pure rotation, would expose how much the prior depends on VO quality.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Here's my take: the core idea is genuinely useful and the ablation is clean, but the paper's headline claim that it 'outperforms existing approaches' rests on a comparison that I don't think is valid as reported. The stress-test note is on target. The test set is a custom 8,691-frame split from KITTI odometry sequences 00/04/05/07, and the competitor numbers in Table 2 are, for Monodepth at least, the exact published numbers from the Eigen 697-frame test. That exact match across a 12x larger and differently distributed set is the kind of thing that happens when a number is copied, not recomputed. The paper says the split is 'fully consistent' with Eigen training (no overlap), which prevents leakage but does not make the two test distributions equivalent. Unless every competitor was rerun on the 8,691-frame split, the 0.091 vs 0.108 comparison is apples-to-oranges. This is load-bearing because it is the entire basis of the SOTA claim.

What I like: the idea of using a sparse, scale-aware VO prior at inference time is simple and practical, and the sparsity-invariant autoencoder is a sensible way to densify a 0.06% point cloud. The ablations in Table 1 are a real strength: each addition (autoencoder, skip, symmetry, fine-tuning) gives a consistent, mostly monotonic improvement, which is more than many papers in this area show. The runtime analysis on TX2 and 2080Ti is also useful and honest about the overhead. The method is architecture-agnostic in the demonstrated sense, working with both Monodepth and PyD-Net.

Soft spots beyond the split issue: the monocular VO pipeline is proprietary (Zenuity) and not described. That matters because the whole method depends on its quality and scale recovery. No code, no released models, no error bars. The inner/outer losses do feed the sparse VO points back into the network, so the evaluation has to rely on LiDAR ground truth; that is independent, so I don't see a circularity problem.

Who is this for: people working on self-supervised depth for robotics/AR, especially those who already have a VO/SLAM front-end. The paper deserves a serious referee, but I'd send it back for a major revision: recompute all comparisons on a common split, release or at least thoroughly describe the VO pipeline, and add variance/error bars.

Referee Report

3 major / 3 minor

Summary. The paper proposes a method for improving self-supervised monocular depth estimation by injecting a sparse geometric prior obtained from a visual odometry (VO) algorithm. A sparsity-invariant autoencoder densifies the sparse VO disparity maps, and the densified map is concatenated with the RGB image and fed to an existing depth estimation network (Monodepth or PyD-Net). Training uses stereo self-supervision plus inner and outer losses that enforce agreement between the network outputs and the sparse VO points. Experiments on KITTI show consistent improvements in an ablation study as components are added, and the paper claims state-of-the-art results, with additional runtime experiments on a 2080 Ti GPU and a Jetson TX2.

Significance. If the results hold, the idea of using a VO-derived sparse prior at inference time (rather than only at training time) is a simple and potentially effective way to improve self-supervised monocular depth estimation. The paper's strengths include a clear ablation that shows monotonic improvement from adding each module, evaluation against LiDAR ground truth, compatibility with both large and compact backbone architectures, and a thorough runtime analysis. The central claim of outperforming existing approaches, however, is undermined by an invalid comparison protocol, and the reliance on an undisclosed proprietary VO pipeline substantially limits reproducibility. The approach is novel in its use of a monocular VO prior at deployment, but the evidence as presented is not sufficient to support the state-of-the-art claim.

major comments (3)

[Section 4.4, Table 2] The claim that VOMonodepth outperforms existing approaches is based on comparing numbers obtained on a custom 8691-frame test split with published results reported on the standard 697-frame Eigen test set. The paper only shows that the 8691 frames do not overlap the Eigen training set, which does not make the two test distributions equivalent; the exact repetition of Monodepth's Abs Rel 0.108 in both Tables 1 and 2 strongly suggests that the competitor value was copied from the original paper rather than recomputed on the new split. To support the SOTA claim, the authors should either re-evaluate all competitor models on the same 8691-frame split or restrict the claim to comparisons on the standard Eigen test set.
[Section 4.2] The monocular VO pipeline provided by Zenuity is used for the final fine-tuning and for all inference, yet this pipeline is not described, characterized, or released. Consequently, the reported gains are conditional on an unavailable black-box component, and the paper's central claim that a traditional visual odometry prior improves depth estimation cannot be independently reproduced with a public VO algorithm. Please provide details of the scale-recovery mechanism and, ideally, repeat the key experiments with an open-source monocular VO baseline.
[Section 4.3, Table 1] All reported results come from a single training run, with no error bars or multiple random seeds. Given that the final method improves over Monodepth-ResNet by small margins on some metrics (e.g., δ < 1.25), the reader cannot assess whether the observed improvements are statistically significant or within run-to-run variation. The authors should report mean and standard deviation over multiple runs, or at least show that the differences are consistent with a paired evaluation.

minor comments (3)

[Equation (5)] The left-right consistency loss is written as |dL - dR + dL|, which simplifies to |2dL - dR|; the standard formulation samples dR at a shifted location, so this appears to be a typesetting error that should be corrected to the usual sampling expression.
[Table 1, Section 4.3] The baseline configuration described as 'directly feeding the depth estimator with VO input' is not fully specified: it is unclear how the sparse VO map is concatenated to the RGB image (e.g., as a fourth input channel) and whether the depth estimator's first convolutional layer was modified to accept the additional input.
[Section 4.2] The list of loss weights contains a naming inconsistency: βapp and βds correspond to Eq. (2), but αlr appears instead of βlr; please align the notation between the text and the equation.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the sparse VO points are an explicit input cue, the dense prediction is evaluated against LiDAR, and the sparse-point agreement losses are the intended training objective rather than a hidden reduction.

full rationale

The paper's derivation chain is: a monocular VO algorithm supplies sparse 3D points, a sparsity-invariant autoencoder densifies them, and the densified prior is concatenated with the RGB image and fed to a self-supervised depth network. The final depth is trained with stereo reprojection, an inner loss (Eq. 7) matching the autoencoder output to the sparse VO disparities at defined pixels, and an outer loss (Eq. 8) matching the final output to the same sparse disparities. If fully optimized, these losses do force agreement with the VO input at the sparse pixels, but this is the explicit design of the method, not a hidden equivalence: the reported metrics are computed against LiDAR ground truth over all valid pixels, the dense output at non-sparse pixels is not determined by the sparse input, and the paper does not claim to predict the VO points as its output. The concern about comparing against prior work on a custom 8691-frame split is a benchmarking validity issue, not circularity. Self-citations (e.g., PyD-Net, 3Net) are used as baselines and architectural components, not as load-bearing justification for the central improvement claim. Therefore the derivation is self-contained with respect to the circularity patterns considered.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim depends on tuned loss weights, an external monocular VO pipeline, and assumptions about the transferability of stereo self-supervision and sparse convolutions.

free parameters (2)

Loss weights (alpha_st, alpha_in, alpha_out, beta_app, beta_ds, alpha_lr, alpha_occ) = 1, 5, 2, 1, 0.1/r, 1, 0.01
Tuned on KITTI to balance the stereo, inner, and outer loss terms; no theory dictates these values.
Autoencoder architecture (kernel sizes 9, 5, 3, 3; 16 filters; stride 1) = Stated in Section 3.1
Manual design choices selected to work with sparse data; not derived from first principles.

assumptions (4)

domain assumption Stereo self-supervision loss from Godard et al. [16] provides a valid training signal.
The entire training relies on this loss to learn depth without ground truth.
domain assumption Sparsity-invariant convolution as in Uhrig et al. [41] correctly handles sparse inputs.
The autoencoder uses sparse convolutions to densify VO points; if this does not generalize, the prior will be corrupted.
domain assumption Monocular VO with scale recovery yields accurate sparse depths at test time.
The paper depends on this external pipeline (Zenuity) for the reported results; it is not described in detail.
domain assumption The custom test split is compliant with the Eigen protocol and enables fair comparison.
They introduce a split of 8691 frames from KITTI odometry; they claim no overlap with Eigen's training set, but this is an untested assumption.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing self-supervised monocular depth estimation with traditional visual odometry." pith.science (2026). https://pith.science/paper/JADJ44Z4

@misc{pith2026190803127,
  author       = {Pith},
  title        = {Pith review of: Enhancing self-supervised monocular depth estimation with traditional visual odometry},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JADJ44Z4}},
  note         = {Machine review of arXiv:1908.03127}
}

read the original abstract

Estimating depth from a single image represents an attractive alternative to more traditional approaches leveraging multiple cameras. In this field, deep learning yielded outstanding results at the cost of needing large amounts of data labeled with precise depth measurements for training. An issue softened by self-supervised approaches leveraging monocular sequences or stereo pairs in place of expensive ground truth depth annotations. This paper enables to further improve monocular depth estimation by integrating into existing self-supervised networks a geometrical prior. Specifically, we propose a sparsity-invariant autoencoder able to process the output of conventional visual odometry algorithms working in synergy with depth-from-mono networks. Experimental results on the KITTI dataset show that by exploiting the geometrical prior, our proposal: i) outperforms existing approaches in the literature and ii) couples well with both compact and complex depth-from-mono architectures, allowing for its deployment on high-end GPUs as well as on embedded devices (e.g., NVIDIA Jetson TX2).

Figures

Figures reproduced from arXiv: 1908.03127 by the authors.

**Figure 2.** Overview of our framework. Sparse depths (SD) provided by a VO algorithm are fed to a sparse auto-encoder producing more [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Structure of the sparsity-invariant autoencoder. Four [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Qualitative results on KITTI dataset. On each column, from top to bottom: reference image, sparse VO points, depth map outputs [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

**Figure 5.** Figure 5: Qualitative comparison on KITTI dataset. (a) input im [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

52 extracted references · 45 canonical work pages

[1]

Aleotti, F

F. Aleotti, F. Tosi, M. Poggi, and S. Mattoccia. Generative adversarial networks for unsupervised monocular depth pre- diction. In 15th European Conference on Computer Vision (ECCV) Workshops, 2018. 2

work page 2018
[2]

Atapour-Abarghouei and T

A. Atapour-Abarghouei and T. P. Breckon. Real-time monocular depth estimation using synthetic data with do- main adaptation via image style transfer. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, volume 18, page 1, 2018. 2

work page 2018
[3]

Casser, S

V . Casser, S. Pirk, R. Mahjourian, and A. Angelova. Un- supervised learning of depth and ego-motion: A structured approach. In Thirty-Third AAAI Conference on Artiﬁcial In- telligence (AAAI-19), 2019. 2, 7

work page 2019
[4]

Chodosh, C

N. Chodosh, C. Wang, and S. Lucey. Deep convolutional compressed sensing for lidar depth completion. InComputer Vision - ACCV 2018 - 14th Asian Conference on Computer Vision, Perth, Australia, December 2-6, 2018, Revised Se- lected Papers, Part I, pages 499–513, 2018. 3

work page 2018
[5]

Cordts, M

M. Cordts, M. Omran, S. Ramos, T. Rehfeld, M. Enzweiler, R. Benenson, U. Franke, S. Roth, and B. Schiele. The cityscapes dataset for semantic urban scene understanding. CoRR, abs/1604.01685, 2016. 6, 7

arXiv 2016
[6]

CS Kumar, S

A. CS Kumar, S. M. Bhandarkar, and P. Mukta. Monocu- lar depth prediction using generative adversarial networks. In 1st International Workshop on Deep Learning for Visual SLAM, (CVPR), 2018. 2

work page 2018
[7]

Eigen, C

D. Eigen, C. Puhrsch, and R. Fergus. Depth map prediction from a single image using a multi-scale deep network.CoRR, abs/1406.2283, 2014. 2, 5, 6, 7

arXiv 2014
[8]

Eldesokey, M

A. Eldesokey, M. Felsberg, and F. S. Khan. Propagat- ing conﬁdences through cnns for sparse data regression. In British Machine Vision Conference 2018, BMVC 2018, Northumbria University, Newcastle, UK, September 3-6, 2018, page 14, 2018. 3

work page 2018

Show all 52 references

[9]

Engel, V

J. Engel, V . Koltun, and D. Cremers. Direct sparse odometry. Mar. 2018. 2

2018
[10]

Engel, T

J. Engel, T. Sch ¨ops, and D. Cremers. LSD-SLAM: Large- scale direct monocular SLAM. September 2014. 2

2014
[11]

Engel, J

J. Engel, J. St ¨uckler, and D. Cremers. Large-scale direct slam with stereo cameras. In 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1935–1942. IEEE, 2015. 2

2015
[12]

Fanani, A

N. Fanani, A. Sturck, M. Barnada, and R. Mester. Multi- modal scale estimation for monocular visual odometry. In IEEE Intelligent Vehicles Symposium, IV 2017, Los Angeles, CA, USA, June 11-14, 2017, pages 1714–1721, 2017. 2, 3, 6

2017
[13]

H. Fu, M. Gong, C. Wang, K. Batmanghelich, and D. Tao. Deep ordinal regression network for monocular depth esti- mation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 1, 2

2018
[14]

R. Garg, V . K. B. G, and I. D. Reid. Unsupervised CNN for single view depth estimation: Geometry to the rescue.CoRR, abs/1603.04992, 2016. 2, 4

2016 arXiv
[15]

A. Geiger. Are we ready for autonomous driving? the kitti vision benchmark suite. In Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recog- nition (CVPR), CVPR ’12, pages 3354–3361, Washington, DC, USA, 2012. IEEE Computer Society. 1, 2, 5, 6, 7

2012
[16]

Godard, O

C. Godard, O. Mac Aodha, and G. J. Brostow. Unsuper- vised monocular depth estimation with left-right consistency. CoRR, abs/1609.03677, 2016. 1, 2, 3, 4, 5, 6, 7, 8

2016 arXiv
[17]

X. Guo, H. Li, S. Yi, J. Ren, and X. Wang. Learning monocu- lar depth by distilling cross-domain stereo networks. In Pro- ceedings of the European Conference on Computer Vision (ECCV), pages 484–500, 2018. 2

2018
[18]

Hartley and A

R. Hartley and A. Zisserman. Multiple View Geometry in Computer Vision. Cambridge University Press, New York, NY , USA, 2 edition, 2003. 1

2003
[19]

K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. InThe IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016. 4

2016
[20]

Hirschmuller

H. Hirschmuller. Stereo processing by semiglobal matching and mutual information. IEEE Transactions on pattern anal- ysis and machine intelligence, 30(2):328–341, 2008. 2, 7

2008
[21]

I. P. Howard. Perceiving in depth, vol. 1: Basic mechanisms
[22]

D. P. Kingma and J. Ba. Adam: A method for stochastic optimization. CoRR, abs/1412.6980, 2014. 6

2014 arXiv
[23]

Klodt and A

M. Klodt and A. Vedaldi. Supervising the new with the old: learning sfm from sfm. In The European Conference on Computer Vision (ECCV), September 2018. 2

2018
[24]

Kuznietsov, J

Y . Kuznietsov, J. St ¨uckler, and B. Leibe. Semi-supervised deep learning for monocular depth map prediction. CoRR, abs/1702.02706, 2017. 2

2017 arXiv
[25]

Ladicky, J

L. Ladicky, J. Shi, and M. Pollefeys. Pulling things out of perspective. In 2014 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014, Columbus, OH, USA, June 23-28, 2014, pages 89–96, 2014. 2

2014
[26]

Laina, C

I. Laina, C. Rupprecht, V . Belagiannis, F. Tombari, and N. Navab. Deeper depth prediction with fully convolutional residual networks. In 3D Vision (3DV), 2016 Fourth Interna- tional Conference on, pages 239–248. IEEE, 2016. 2

2016
[27]

F. Liu, C. Shen, G. Lin, and I. Reid. Learning depth from sin- gle monocular images using deep convolutional neural ﬁelds. IEEE transactions on pattern analysis and machine intelli- gence, 38(10):2024–2039, 2016. 2

2024
[28]

Y . Luo, J. Ren, M. Lin, J. Pang, W. Sun, H. Li, and L. Lin. Single view stereo matching. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 2

2018
[29]

Ma and S

F. Ma and S. Karaman. Sparse-to-dense: Depth prediction from sparse depth samples and a single image. In 2018 IEEE International Conference on Robotics and Automation, ICRA 2018, Brisbane, Australia, May 21-25, 2018, pages 1– 8, 2018. 2, 3

2018
[30]

Mahjourian, M

R. Mahjourian, M. Wicke, and A. Angelova. Unsupervised learning of depth and ego-motion from monocular video using 3d geometric constraints. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , June
[31]

Mur-Artal and J

R. Mur-Artal and J. D. Tard ´os. ORB-SLAM2: an open- source SLAM system for monocular, stereo and RGB-D cameras. CoRR, abs/1610.06475, 2016. 2, 3, 6

2016 arXiv
[32]

R. A. Newcombe, S. J. Lovegrove, and A. J. Davison. Dtam: Dense tracking and mapping in real-time. In 2011 inter- national conference on computer vision , pages 2320–2327. IEEE, 2011. 2

2011
[33]

Peluso, A

V . Peluso, A. Cipolletta, A. Calimera, M. Poggi, F. Tosi, and S. Mattoccia. Enabling energy-efﬁcient unsupervised monocular depth estimation on armv7-based platforms. In Design Automation and Test in Europe (DATE), 2019. 2

2019
[34]

Pilzer, S

A. Pilzer, S. Lathuiliere, N. Sebe, and E. Ricci. Reﬁne and distill: Exploiting cycle-inconsistency and knowledge distil- lation for unsupervised monocular depth estimation. In The IEEE Conference on Computer Vision and Pattern Recogni- tion (CVPR), 2019. 2, 6

2019
[35]

Poggi, F

M. Poggi, F. Aleotti, F. Tosi, and S. Mattoccia. Towards real-time unsupervised monocular depth estimation on cpu. In IEEE/JRS Conference on Intelligent Robots and Systems (IROS), 2018. 2, 4, 6, 7, 8

2018
[36]

Poggi, F

M. Poggi, F. Tosi, and S. Mattoccia. Learning monocular depth estimation with unsupervised trinocular assumptions. In 6th International Conference on 3D Vision (3DV) , 2018. 2, 7

2018
[37]

Saxena, M

A. Saxena, M. Sun, and A. Y . Ng. Make3d: Learning 3d scene structure from a single still image.IEEE Trans. Pattern Anal. Mach. Intell., 31(5):824–840, May 2009. 2

2009
[38]

Shen and S.-C

J. Shen and S.-C. S. Cheung. Layer depth denoising and completion for structured-light rgb-d cameras. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2013. 2

2013
[39]

Tateno, F

K. Tateno, F. Tombari, I. Laina, and N. Navab. Cnn-slam: Real-time dense monocular slam with learned depth predic- tion. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 6243–6252, 2017. 2

2017
[40]

F. Tosi, F. Aleotti, M. Poggi, and S. Mattoccia. Learn- ing monocular depth estimation infusing traditional stereo knowledge. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2019. 2, 4, 7

2019
[41]

Uhrig, N

J. Uhrig, N. Schneider, L. Schneider, U. Franke, T. Brox, and A. Geiger. Sparsity invariant cnns. CoRR, abs/1708.06500,

arXiv
[42]

C. Wang, J. Miguel Buenaposada, R. Zhu, and S. Lucey. Learning depth from monocular videos using direct meth- ods. In The IEEE Conference on Computer Vision and Pat- tern Recognition (CVPR), June 2018. 2, 7

2018
[43]

X. Wang, H. Zhang, X. Yin, M. Du, and Q. Chen. Monocular visual odometry scale recovery using geometrical constraint. In 2018 IEEE International Conference on Robotics and Au- tomation (ICRA), pages 988–995, May 2018. 2, 3, 6

2018
[44]

Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simon- celli. Image quality assessment: from error visibility to structural similarity. IEEE transactions on image process- ing, 13(4):600–612, 2004. 4

2004
[45]

D. Xu, W. Wang, H. Tang, H. Liu, N. Sebe, and E. Ricci. Structured attention guided convolutional neural ﬁelds for monocular depth estimation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 2

2018
[46]

N. Yang, R. Wang, J. St ¨uckler, and D. Cremers. Deep vir- tual stereo odometry: Leveraging deep depth prediction for monocular direct sparse odometry. In European Conference on Computer Vision, pages 835–852. Springer, 2018. 2, 5, 6

2018
[47]

X. Yin, X. Wang, X. Du, and Q. Chen. Scale recovery for monocular visual odometry using depth estimated with deep convolutional neural ﬁelds. InProceedings of the IEEE Inter- national Conference on Computer Vision, pages 5870–5878,
[48]

Yin and J

Z. Yin and J. Shi. Geonet: Unsupervised learning of dense depth, optical ﬂow and camera pose. In The IEEE Confer- ence on Computer Vision and Pattern Recognition (CVPR) , June 2018. 2, 7

2018
[49]

Zama Ramirez, M

P. Zama Ramirez, M. Poggi, F. Tosi, S. Mattoccia, and L. Di Stefano. Geometry meets semantic for semi-supervised monocular depth estimation. In 14th Asian Conference on Computer Vision (ACCV), 2018. 2

2018
[50]

H. Zhan, R. Garg, C. Saroj Weerasekera, K. Li, H. Agarwal, and I. Reid. Unsupervised learning of monocular depth es- timation and visual odometry with deep feature reconstruc- tion. In The IEEE Conference on Computer Vision and Pat- tern Recognition (CVPR), June 2018. 2

2018
[51]

T. Zhou, M. Brown, N. Snavely, and D. G. Lowe. Unsuper- vised learning of depth and ego-motion from video. In The IEEE Conference on Computer Vision and Pattern Recogni- tion (CVPR), July 2017. 1, 2, 7

2017
[52]

Y . Zou, Z. Luo, and J.-B. Huang. Df-net: Unsupervised joint learning of depth and ﬂow using cross-task consistency. In European Conference on Computer Vision, 2018. 2, 7

2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Aleotti, F

F. Aleotti, F. Tosi, M. Poggi, and S. Mattoccia. Generative adversarial networks for unsupervised monocular depth pre- diction. In 15th European Conference on Computer Vision (ECCV) Workshops, 2018. 2

work page 2018

[2] [2]

Atapour-Abarghouei and T

A. Atapour-Abarghouei and T. P. Breckon. Real-time monocular depth estimation using synthetic data with do- main adaptation via image style transfer. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, volume 18, page 1, 2018. 2

work page 2018

[3] [3]

Casser, S

V . Casser, S. Pirk, R. Mahjourian, and A. Angelova. Un- supervised learning of depth and ego-motion: A structured approach. In Thirty-Third AAAI Conference on Artiﬁcial In- telligence (AAAI-19), 2019. 2, 7

work page 2019

[4] [4]

Chodosh, C

N. Chodosh, C. Wang, and S. Lucey. Deep convolutional compressed sensing for lidar depth completion. InComputer Vision - ACCV 2018 - 14th Asian Conference on Computer Vision, Perth, Australia, December 2-6, 2018, Revised Se- lected Papers, Part I, pages 499–513, 2018. 3

work page 2018

[5] [5]

Cordts, M

M. Cordts, M. Omran, S. Ramos, T. Rehfeld, M. Enzweiler, R. Benenson, U. Franke, S. Roth, and B. Schiele. The cityscapes dataset for semantic urban scene understanding. CoRR, abs/1604.01685, 2016. 6, 7

arXiv 2016

[6] [6]

CS Kumar, S

A. CS Kumar, S. M. Bhandarkar, and P. Mukta. Monocu- lar depth prediction using generative adversarial networks. In 1st International Workshop on Deep Learning for Visual SLAM, (CVPR), 2018. 2

work page 2018

[7] [7]

Eigen, C

D. Eigen, C. Puhrsch, and R. Fergus. Depth map prediction from a single image using a multi-scale deep network.CoRR, abs/1406.2283, 2014. 2, 5, 6, 7

arXiv 2014

[8] [8]

Eldesokey, M

A. Eldesokey, M. Felsberg, and F. S. Khan. Propagat- ing conﬁdences through cnns for sparse data regression. In British Machine Vision Conference 2018, BMVC 2018, Northumbria University, Newcastle, UK, September 3-6, 2018, page 14, 2018. 3

work page 2018