</> Code with Papers by Syntology

Browse › Computer Vision › Monocular Depth Estimation › KITTI Eigen split

Monocular Depth Estimation archive 2025-07-28

KITTI Eigen split Benchmark (Monocular Depth Estimation)

79 rows 68 with code listed 8 metrics Dataset page

Monocular Depth Estimation is the task of estimating the depth value (distance relative to the camera) of each pixel given a single (monocular) RGB image. This challenging task is a key prerequisite for determining scene understanding for applications such as 3D scene reconstruction, autonomous driving, and AR. State-of-the-art methods usually fall into one of two categories: designing a complex network that is powerful enough to directly regress the depth map, or splitting the input into bins or windows to reduce computational complexity. The most popular benchmarks are the KITTI and NYUv2 datasets. Models are typically evaluated using RMSE or absolute relative error.

Source: Defocus Deblurring Using Dual-Pixel Data

The archive carries no text for this table; the description above is the archive's text for the task Monocular Depth Estimation. archive 2025-07-28

Over time archive 2025-07-28

The chart needs JavaScript; the table below carries every value.

Direction inferred from the metric name, not from the archive: absolute relative error (lower is better), RMSE (lower is better), RMSE log (lower is better), Square relative error (SqRel) (lower is better). Not inferred (points only, no best-so-far line): Sq Rel, Delta < 1.25, Delta < 1.25^2, Delta < 1.25^3. Points are placed at the row's paper date; 79 of 79 rows carry one.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 SPIDepth 0.0291.3940.0690.0480.990.9991.000 ✓ Paper Code 2024 13 of 14 ran · 1 unverified report
2 UniK3D (FT, metric) 0.0371.680.0600.9900.9980.999 – Paper Code 2025 linked, not harvested report
3 UniDepthV2 (FT, metric) 0.0371.710.0610.9890.9980.999 ✓ Paper Code 2025 2 of 2 ran · 0 unverified report
4 Metric3Dv2 (g2, FT, 80m, flip_aug_test) 0.0391.7660.0600.9890.9981.000 ✓ Paper Code 2024 3 of 3 ran · 0 unverified report
5 LightedDepth (Video Method) 0.0411.7480.1070.0590.9890.9980.999 – Paper Code 2023 linked, not harvested report
6 FutureDepth 0.0411.8560.1170.0660.9840.9981.0000.117 – Paper – 2024 no code linked report
7 UniDepth (Zero-shot) 0.0421.750.0640.9860.9980.999 ✓ Paper Code 2024 15 of 26 ran · 11 unverified report
8 SQLdepth (ConvNeXt-L) 0.0431.6980.1050.0640.9830.9980.999 – Paper Code 2023 linked, not harvested report
9 AFNet 0.0441.7120.1320.0690.9800.9970.999 – Paper Code 2024 7 of 8 ran · 1 unverified report
10 Depth Anything 0.0461.8960.1210.0690.9820.9981.000 ✓ Paper Code 2024 3 of 11 ran · 8 unverified report
11 MetaPrompt-SD 0.0471.9280.1250.0710.9810.9981.000 ✓ Paper Code 2023 4 of 6 ran · 2 unverified report
12 ECoDepth 0.0481.9660.1390.0740.9790.9981.000 – Paper Code 2024 11 of 15 ran · 4 unverified report
13 ScaleDepth-K 0.0481.9870.1360.0730.980.9981.000 – Paper Code 2024 linked, not harvested report
14 EVP 0.0482.0150.1360.0730.9800.9981.000 – Paper Code 2023 linked, not harvested report
15 GEDepth 0.0482.0440.1420.0760.97630.99720.9993 – Paper Code 2023 linked, not harvested report
16 MAMo 0.0491.9840.130.0720.9770.9980.9995 – Paper – 2023 no code linked report
17 SwinV2-L 1K-MIM 0.0501.9660.1390.0750.9770.9981.000 – Paper Code 2022 4 of 5 ran · 1 unverified report
18 IEBins 0.0502.0110.1420.0750.9780.9980.999 – Paper Code 2023 8 of 13 ran · 5 unverified report
19 NDDepth 0.0502.0250.1410.0750.9780.9980.999 – Paper Code 2023 linked, not harvested report
20 URCDC-Depth 0.0502.0320.1420.0760.9770.9970.999 – Paper Code 2023 5 of 11 ran · 6 unverified report
21 iDisc 0.0502.0670.1450.0770.9770.9970.999 – Paper Code 2023 5 of 6 ran · 1 unverified report
22 DDP (Swin-L, step-3) 0.0502.0720.1480.0760.9750.9970.999 – Paper Code 2023 linked, not harvested report
23 MIM-Swin-V2 0.05082.03730.14580.0770.97570.99740.9994 – Paper – 2023 no code linked report
24 PixelFormer 0.0512.0810.1490.0770.9760.9970.999 ✓ Paper Code 2022 5 of 11 ran · 6 unverified report
25 SwinV2-B 1K-MIM 0.0522.0500.1480.0780.9760.9980.999 – Paper Code 2022 4 of 5 ran · 1 unverified report
26 BinsFormer 0.0522.0980.1510.0790.9740.9970.999 ✓ Paper Code 2022 1 of 5 ran · 4 unverified report
27 NeWCRFs 0.0522.1290.1550.0790.9740.9970.999 ✓ Paper Code 2022 5 of 10 ran · 5 unverified report
28 DepthFormer 0.0522.1430.1580.0790.9750.9970.999 ✓ Paper Code 2022 linked, not harvested report
29 MonoDELSNet 0.0532.1010.1610.0820.9690.9960.999 – Paper Code 2021 linked, not harvested report
30 SfM-Revisited 0.0552.2730.224 – Paper Code 2021 5 of 21 ran · 16 unverified report
31 D-Net 0.0562.3620.1890.0870.9630.9950.999 – Paper Code 2021 linked, not harvested report
32 GLPDepth 0.0572.2970.0860.9670.9960.999 – Paper Code 2022 1 of 2 ran · 1 unverified report
33 NVS-MonoDepth 0.0572.702 – Paper – 2021 no code linked report
34 Depthformer 0.0582.2850.187 0.0870.9670.9960.999 ✓ Paper Code 2022 1 of 2 ran · 1 unverified report
35 AdaBins 0.0582.3600.0880.9640.9950.999 ✓ Paper Code 2020 linked, not harvested report
36 Metric3D (zero-shot) 0.0582.770.9670.9950.999 – Paper Code 2023 linked, not harvested report
37 LapDepth 0.0592.4460.0910.9620.9940.999 – Paper Code 2021 linked, not harvested report
38 DPT-Hybrid 0.0622.5730.0920.9590.9950.999 – Paper Code 2021 54 of 116 ran · 62 unverified report
39 BTS 0.064 – Paper Code 2019 5 of 15 ran · 10 unverified report
40 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) 0.06522.11280.17970.08820.9680.9970.9993 – Paper Code 2023 21 of 46 ran · 25 unverified report
41 LightDepth 0.0702.923 – Paper Code 2022 linked, not harvested report
42 DORN 0.0722.7270.1200.9320.9840.994 – Paper Code 2018 0 of 3 ran · 3 unverified report
43 VNL 0.072 – Paper Code 2019 linked, not harvested report
44 PrimeDepth + Depth Anything 0.0730.953 – Paper Code 2024 13 of 15 ran · 2 unverified report
45 DSN 0.075 – Paper – 2020 no code linked report
46 PrimeDepth 0.0790.937 – Paper Code 2024 13 of 15 ran · 2 unverified report
47 Focal-WNet 0.0823.0760.1200.9260.9860.997 – Paper Code 2022 linked, not harvested report
48 DepthMaster 0.0820.937 ✓ Paper Code 2025 linked, not harvested report
49 Manydepth2 0.0914.2320.1700.6490.9090.9680.984 – Paper Code 2023 0 of 9 ran · 9 unverified report
50 VOMonodepth 0.091 – Paper – 2019 no code linked report
51 DenseDepth 0.093 – Paper Code 2018 5 of 23 ran · 18 unverified report
52 SVS 0.094 – Paper Code 2018 linked, not harvested report
53 monoResMatch 0.096 – Paper Code 2019 linked, not harvested report
54 SemiDepth 0.096 – Paper Code 2019 linked, not harvested report
55 CFA 0.096 – Paper – 2018 no code linked report
56 Depth Hints 0.096 – Paper Code 2019 2 of 2 ran · 0 unverified report
57 SOM 0.097 – Paper – 2019 no code linked report
58 Marigold 0.0993.3040.1380.9160.9870.996 ✓ Paper Code 2023 14 of 26 ran · 12 unverified report
59 GCNDepth 0.1044.4940.1810.8880.9650.984 – Paper Code 2021 linked, not harvested report
60 monodepth2 M 0.106 ✓ Paper Code 2018 17 of 24 ran · 7 unverified report
61 DeepLabV3+ (F10) 0.110 – Paper Code 2020 linked, not harvested report
62 DiPE 0.112 – Paper Code 2020 linked, not harvested report
63 DNet 0.1134.8120.1910.8770.9600.981 – Paper Code 2020 2 of 2 ran · 0 unverified report
64 LSIM 0.113 – Paper Code 2019 linked, not harvested report
65 SC-Depth (ResNet 50) 0.1144.7060.1910.8730.9600.982 – Paper Code 2021 linked, not harvested report
66 SemanticAware 0.118 – Paper – 2019 no code linked report
67 SC-Depth (ResNet18) 0.1194.9500.1970.8630.9570.981 – Paper Code 2021 linked, not harvested report
68 PackNet-SfM 0.12 – Paper Code 2019 linked, not harvested report
69 3Net 0.126 – Paper Code 2018 linked, not harvested report
70 SC-SfMLearner_CS+K 0.128 – Paper Code 2019 2 of 3 ran · 1 unverified report
71 SelfDepthNorm 0.133 – Paper – 2019 no code linked report
72 SIGNet 0.133 – Paper Code 2018 linked, not harvested report
73 struct2depth 0.135 – Paper Code 2018 1 of 3 ran · 2 unverified report
74 SC-SfMLearner 0.137 – Paper Code 2019 2 of 3 ran · 1 unverified report
75 CC 0.140 – Paper Code 2018 1 of 14 ran · 13 unverified report
76 SIW 0.14 – Paper – 2022 no code linked report
77 LeReS 0.1490.784 – Paper Code 2020 linked, not harvested report
78 GASDA 0.149 – Paper Code 2019 linked, not harvested report
79 VDA 0.193 – Paper Code 2019 linked, not harvested report

All 79 rows shown. 79 link to a paper page on this site; 15 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 35 rows have a graph line, from 32 distinct papers; 33 rows (30 papers) have at least one sample that ran. Counting each paper once: Syntology ran 235 of 472 samples; 237 unverified. Separately, 121 of those 472 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.

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