Papers › Digging Into Self-Supervised Monocular Depth Estimation

Digging Into Self-Supervised Monocular Depth Estimation

4 Jun 2018arXiv:1806.01260archive 2025-07-28

Clément Godard, Oisin Mac Aodha, Michael Firman, Gabriel Brostow

Per-pixel ground-truth depth data is challenging to acquire at scale. To overcome this limitation, self-supervised learning has emerged as a promising alternative for training models to perform monocular depth estimation. In this paper, we propose a set of improvements, which together result in both quantitatively and qualitatively improved depth maps compared to competing self-supervised methods. Research on self-supervised monocular training usually explores increasingly complex architectures, loss functions, and image formation models, all of which have recently helped to close the gap with fully-supervised methods. We show that a surprisingly simple model, and associated design choices, lead to superior predictions. In particular, we propose (i) a minimum reprojection loss, designed to robustly handle occlusions, (ii) a full-resolution multi-scale sampling method that reduces visual artifacts, and (iii) an auto-masking loss to ignore training pixels that violate camera motion assumptions. We demonstrate the effectiveness of each component in isolation, and show high quality, state-of-the-art results on the KITTI benchmark.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

The archive's official repository (nianticlabs/monodepth2) was not harvested; the 24 samples below come from 5 community repositories.

Syntology 24 code samples harvested from 5 repositories linked to this paper: 23 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 4 honoured, 1 violated, 8 with no contract checked; 10 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; 4 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 24 samples from 5 repositories, 23 ran (13 with no instrument failure; 10 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 23 samples that ran has no recorded date.

15 repositories listed; official and paper-mentioned ones first.

nianticlabs/monodepth2 officialmentioned in papermentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
CaptainEven/MonoDepthV2 mentioned on GitHubpytorchSyntology: 4 samples, 4 ran (3 instrument-failed) · 0 unverified report
FangGet/tf-monodepth2 mentioned on GitHubtfSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
IcarusWizard/monodepth2-paddle mentioned on GitHubpaddleApache-2.0Syntology: 8 samples, 8 ran (3 instrument-failed) · 0 unverified report
NiallEHunt/MonocularDepth-Using-LightFields mentioned on GitHubpytorchSyntology: not harvested report
TWJianNuo/panoptic-scene-understanding mentioned on GitHubpytorchSyntology: not harvested report
TanyaChutani/Monodepth-Tf2.x mentioned on GitHubtfSyntology: not harvested report
XXXVincent/MonoDepth2 mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
iodncookie/my_loss_monodepth_master mentioned on GitHubtfSyntology: 2 samples, 2 ran (1 instrument-failed) · 0 unverified report
isennkubilay/monodepth2_tf mentioned on GitHubtfSyntology: not harvested report
jzwqaq/monodepth_jzw mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
minghanz/DepthC3D mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
qrzyang/pseudo-stereo mentioned on GitHubpytorchApache-2.0Syntology: 9 samples, 8 ran (3 instrument-failed) · 1 unverified report
rnlee1998/SRD mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
tudelft/filled-disparity-monodepth mentioned on GitHubtfNOASSERTIONSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

24 samples harvested: 23 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 4 honoured, 1 violated, 8 with no contract checked; 10 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

4ran · honoured contract
1ran · violated contract
6ran · our draft was wrong
4ran; the contract check could not run
8ran
1unverified

Licence: 10 of the 24 samples are pointer only, meaning Syntology does not serve that copy's text. For 4 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

CaptainEven/MonoDepthV2 community (archive-listed) 4 samples, 4 ran (3 instrument-failed) · 0 unverified
compute_errors in 41 papers evaluate_depth.py ran · honoured contract fingerprinted licence not identified · pointer only · d4eb240de6e155d3 · report fixture: signature gt:ndarray/3/float, pred:ndarray/3/float · returned tuple · level recorded 2026-09-22
depth2disp test_simple.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence not identified · pointer only · dafc01865fd0e64d · report fixture: signature b:float/0/float, f:float/0/float, depth:ndarray/2/float · returned ndarray · level recorded 2026-09-22
disp2depth test_simple.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence not identified · pointer only · 2f583364067a6299 · report fixture: signature b:float/0/float, f:float/0/float, disp:ndarray/2/float · returned ndarray · level recorded 2026-09-22
project_depth_to_pointcloud test_simple.py ran · our draft was wrong licence not identified · pointer only · 7d5438c1a2189f3f · report fixture: signature calib:StandInCalib/0/?, depth:ndarray/2/float, max_height:float/0/float · returned ndarray · level recorded 2026-09-22
FangGet/tf-monodepth2 community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
concat_image_seq data/prepare_train_data.py ran · violated contract fingerprinted MIT (permissive) · 09d614ea6e098976 · report fixture: signature seq:list/1/ndarray · returned ndarray · level recorded 2026-09-22
IcarusWizard/monodepth2-paddle community (archive-listed) 8 samples, 8 ran (3 instrument-failed) · 0 unverified
batch_post_process_disparity in 21 papers evaluate_depth.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · a8814cc5c348b987 · report fixture: signature l_disp:ndarray/3/float, r_disp:ndarray/3/float · returned ndarray · level recorded, date not recorded
compute_errors in 41 papers evaluate_depth.py ran · honoured contract fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · edd4f86e8f02732f · report fixture: signature gt:ndarray/2/float, pred:ndarray/2/float · returned tuple · level recorded 2026-09-22
disp_to_depth in 21 papers layers.py ran fingerprinted Apache-2.0 (permissive) · 212db0f1a51698c5 · report fixture: signature disp:Tensor/2/float, min_depth:float/0/float, max_depth:float/0/float · returned tuple · level recorded 2026-09-26
load_velodyne_points in 8 papers kitti_utils.py ran · honoured contract Apache-2.0 (permissive) · 8bfc895e86bfc7bc · report fixture: signature filename:str/0/str · returned ndarray · level recorded 2026-09-24
normalize_image in 28 papers utils.py ran fingerprinted Apache-2.0 (permissive) · 8749078cf4be8f13 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-26
read_calib_file in 14 papers kitti_utils.py ran · our draft was wrong Apache-2.0 (permissive) · ff833c099a80a327 · report fixture: signature path:str/0/str · returned dict · level recorded 2026-09-24
readlines in 11 papers utils.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · 859a6ec5fa262fcb · report fixture: signature filename:str/0/str · returned list · level recorded 2026-09-24
sub2ind in 11 papers kitti_utils.py ran Apache-2.0 (permissive) · 04e9b96b63844176 · report fixture: signature matrixSize:tuple/1/int, rowSub:int/0/int, colSub:int/0/int · returned int · level recorded 2026-09-24
iodncookie/my_loss_monodepth_master community (archive-listed) 2 samples, 2 ran (1 instrument-failed) · 0 unverified
count_text_lines monodepth_main.py ran · honoured contract licence not identified · pointer only · bed135371a969180 · report fixture: signature file_path:str/0/str · returned int · level recorded 2026-09-22
post_process_disparity monodepth_main.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence not identified · pointer only · 92cb86f70b5d6123 · report fixture: signature disp:ndarray/3/float · returned ndarray · level recorded 2026-09-22
qrzyang/pseudo-stereo community (archive-listed) 9 samples, 8 ran (3 instrument-failed) · 1 unverified
compute_metric_for_each_image utils/metrics.py ran Apache-2.0 (permissive) · 6d968ec37ba07def · report fixture: signature metric_func:callable/0/? · returned function · level recorded 2026-09-26
convbn in 26 papers models/psm_submodule.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · 15ab3c76823aef7e · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, kernel_size:int/0/int, stride:int/0/int, pad:int/0/int, dilation:int/0/int · returned Sequential · level recorded 2026-09-22
convbn_3d in 20 papers models/psm_submodule.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 46674adf87c5459d · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, kernel_size:int/0/int, stride:int/0/int, pad:int/0/int · returned Sequential · level recorded 2026-09-24
get_smooth_loss losses/utils.py ran fingerprinted Apache-2.0 (permissive) · ee1c671a6243bdf3 · report fixture: signature disp:Tensor/4/float, img:Tensor/4/float, mask:Tensor/4/float · returned Tensor · level recorded 2026-09-26
make_iterative_func utils/experiment.py ran · our draft was wrong Apache-2.0 (permissive) · 8233a45ac266ffd4 · report fixture: signature func:callable/0/? · returned function · level recorded 2026-09-24
make_nograd_func in 17 papers utils/experiment.py ran · our draft was wrong Apache-2.0 (permissive) · c2cc757b82342e5f · report fixture: signature func:callable/0/? · returned function · level recorded 2026-09-24
pfm_imread in 5 papers datasets/data_io.py ran Apache-2.0 (permissive) · c35f13ea74c6be42 · report fixture: signature filename:str/0/str · returned tuple · level recorded 2026-09-25
read_all_lines in 6 papers datasets/data_io.py ran Apache-2.0 (permissive) · e711a03e74028ed6 · report fixture: signature filename:str/0/str · returned list · level recorded 2026-09-25
tensor2float in 15 papers utils/experiment.py unverified Apache-2.0 (permissive) · 198cd80d23544257 · report

Tasks

Camera Pose EstimationDepth EstimationImage ReconstructionMonocular Depth EstimationMotion EstimationScene UnderstandingSelf-Supervised LearningUnsupervised Monocular Depth Estimation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Camera Pose Estimation KITTI Odometry Benchmark Monodepth2 Absolute Trajectory Error [m] 93.04 #7 of 7 Archive leaderboard report
Camera Pose Estimation KITTI Odometry Benchmark Monodepth2 Average Rotational Error er[%] 20.72 #7 of 7 Archive leaderboard report
Camera Pose Estimation KITTI Odometry Benchmark Monodepth2 Average Translational Error et[%] 43.21 #7 of 7 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split monodepth2 M absolute relative error 0.106 #60 of 79 Archive leaderboard report
Monocular Depth Estimation Make3D Monodepth2 Abs Rel 0.322 #2 of 6 Archive leaderboard report
Monocular Depth Estimation Make3D Monodepth2 RMSE 7.417 #2 of 6 Archive leaderboard report
Monocular Depth Estimation Make3D Monodepth2 Sq Rel 3.589 #2 of 6 Archive leaderboard report
Monocular Depth Estimation Mid-Air Dataset Monodepth2 Abs Rel 0.717 #6 of 6 Archive leaderboard report
Monocular Depth Estimation Mid-Air Dataset Monodepth2 RMSE 74.552 #6 of 6 Archive leaderboard report
Monocular Depth Estimation Mid-Air Dataset Monodepth2 RMSE log 0.882 #6 of 6 Archive leaderboard report
Monocular Depth Estimation Mid-Air Dataset Monodepth2 SQ Rel 37.164 #6 of 6 Archive leaderboard report
Monocular Depth Estimation VA (Virtual Apartment) MonoDepth2 Absolute relative error (AbsRel) 0.203 #3 of 3 Archive leaderboard report
Monocular Depth Estimation VA (Virtual Apartment) MonoDepth2 Log root mean square error (RMSE_log) 0.251 #3 of 3 Archive leaderboard report
Monocular Depth Estimation VA (Virtual Apartment) MonoDepth2 Mean average error (MAE) 0.295 #3 of 3 Archive leaderboard report
Monocular Depth Estimation VA (Virtual Apartment) MonoDepth2 Root mean square error (RMSE) 0.432 #3 of 3 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections