CoChat
Sign Up
Cited by 680Advanced Vision and ImagingRobotics and Sensor-Based LocalizationAdvanced Neural Network Applications

Semi-Supervised Deep Learning for Monocular Depth Map Prediction

Kuznietsov, Yevhen, Stuckler, Jorg, Leibe, Bastian·2017·2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
DOI 10.1109/cvpr.2017.238
Open this paper in CoChat — free account

Abstract

Supervised deep learning often suffers from the lack of sufficient training data. Specifically in the context of monocular depth map prediction, it is barely possible to determine dense ground truth depth images in realistic dynamic outdoor environments. When using LiDAR sensors, for instance, noise is present in the distance measurements, the calibration between sensors cannot be perfect, and the measurements are typically much sparser than the camera images. In this paper, we propose a novel approach to depth map prediction from monocular images that learns in a semi-supervised way. While we use sparse ground-truth depth for supervised learning, we also enforce our deep network to produce photoconsistent dense depth maps in a stereo setup using a direct image alignment loss. In experiments we demonstrate superior performance in depth map prediction from single images compared to the state-of-the-art methods.

Integrity signals

Registry

Crossref · 10.1109/cvpr.2017.238

Cites

27 of 27 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

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?

  • Progressive expansion and contraction in a fully invertible manner, without discarding uninformative input variability, increases performance across many tasks.

    “In fact, progressive expansion and contraction in a fully invertible manner, without discarding”

    From 3D Packing for Self-Supervised Monocular Depth Estimation (2020) · cites this paper as [11]

  • Reflection padding significantly reduces border artifacts found in existing approaches.

    “We found that this significantly reduces the border artifacts found in existing approaches, e.g.[15].”

    From Digging Into Self-Supervised Monocular Depth Estimation (2019) · cites this paper as [15]

  • The method outperforms all methods that do not use depth supervision on Make3D.

    “We outperform all methods that do not use depth supervision, with the evaluation criteria from [15].”

    From Digging Into Self-Supervised Monocular Depth Estimation (2019) · cites this paper as [15]

  • Stereo-based training has known scale, so no additional scaling is required during evaluation.

    “This is in contrast to stereo based training where the scale is known and as a result no additional scaling is required during the evaluation e.g.[12,15].”

    From Digging Into Self-Supervised Monocular Depth Estimation (2019) · cites this paper as [15]

  • Post-processing, introduced by [15], is a technique to improve test time results on stereo-trained monocular depth estimation methods by running each test image through the network twice, once unflipped and then flipped.

    “estimation methods by running each test image through the network twice, once unflipped and then flipped.”

    From Digging Into Self-Supervised Monocular Depth Estimation (2019) · cites this paper as [15]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Kuznietsov, Y., Stuckler, J., & Leibe, B.
Affiliations
RWTH Aachen University (DE)
ORCID
Leibe, Bastian
Venue
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2017-07
Pages
2215-2223
ISSN
1063-6919 (ISSN-L)
DOI
10.1109/cvpr.2017.238
OpenAlex
W2593414960
Cited by
680
References
27
Citations by year
2026: 8 · 2025: 22 · 2024: 58 · 2023: 65 · 2022: 75 · 2021: 116 · 2020: 128 · 2019: 139 · 2018: 61 · 2017: 8
FWCI
26.86
Type
conference paper
Language
en
Open access
closed
Locations
  • IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
  • OPUS (Augsburg University) (submittedVersion)
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 › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
Keywords
monocular depth estimation, semi-supervised learning, dense depth maps, LiDAR-camera calibration, outdoor environments

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