</> Code with Papers by Syntology

Browse › Computer Vision › Monocular Depth Estimation › NYU-Depth V2

Monocular Depth Estimation archive 2025-07-28

NYU-Depth V2 Benchmark (Monocular Depth Estimation)

85 rows 71 with code listed 6 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). Not inferred (points only, no best-so-far line): log 10, Delta < 1.25, Delta < 1.25^2, Delta < 1.25^3. Points are placed at the row's paper date; 85 of 85 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 HybridDepth 0.0260.1280.9881.0001.000 – Paper Code 2024 6 of 7 ran · 1 unverified report
2 Distill Any Depth 0.0430.981 – Paper Code 2025 linked, not harvested report
3 UniK3D (FT, metric) 0.0440.1730.0190.9890.9981.000 – Paper Code 2025 linked, not harvested report
4 UniDepthV2 (FT, metric) 0.0460.1800.0200.9880.9981.000 ✓ Paper Code 2025 2 of 2 ran · 0 unverified report
5 PrimeDepth + Depth Anything 0.0460.977 – Paper Code 2024 13 of 15 ran · 2 unverified report
6 Metric3Dv2(L, FT) 0.0470.1830.0200.9890.9981.000 ✓ Paper Code 2024 3 of 3 ran · 0 unverified report
7 DepthMaster 0.0500.972 ✓ Paper Code 2025 linked, not harvested report
8 GRIN 0.0510.251 ✓ Paper – 2024 no code linked report
9 Marigold + E2E FT(zero-shot) 0.0520.966 – Paper Code 2024 16 of 17 ran · 1 unverified report
10 Marigold 0.0550.2240.0240.9640.9910.998 ✓ Paper Code 2023 14 of 26 ran · 12 unverified report
11 Depth Anything 0.0560.2060.0240.9840.9981.000 ✓ Paper Code 2024 3 of 11 ran · 8 unverified report
12 UniDepth (Zero-shot) 0.0580.2010.0240.9840.9970.999 ✓ Paper Code 2024 15 of 26 ran · 11 unverified report
13 PrimeDepth 0.0580.966 – Paper Code 2024 13 of 15 ran · 2 unverified report
14 ECoDepth 0.0590.2180.0260.9780.9970.999 – Paper Code 2024 11 of 15 ran · 4 unverified report
15 MetaPrompt-SD 0.0610.2230.0270.9760.9970.999 ✓ Paper Code 2023 4 of 6 ran · 2 unverified report
16 EVP 0.0610.2240.0270.9760.9970.999 – Paper Code 2023 linked, not harvested report
17 TADP 0.0620.2250.0270.9760.9970.999 ✓ Paper Code 2023 linked, not harvested report
18 FutureDepth 0.0630.2330.0270.9810.9960.999 – Paper – 2024 no code linked report
19 MeSa 0.0660.2380.0290.9640.9950.999 – Paper – 2023 no code linked report
20 PolyMaX(ConvNeXt-L) 0.0670.250.0290.9690.99580.999 ✓ Paper Code 2023 3 of 3 ran · 0 unverified report
21 VPD 0.0690.2540.0300.9640.9950.999 – Paper Code 2023 4 of 5 ran · 1 unverified report
22 NVDS(DPT-L) 0.0720.2820.0310.94930.9910.997 ✓ Paper Code 2023 linked, not harvested report
23 DMD 0.0720.2960.0310.9530.9890.996 ✓ Paper – 2023 no code linked report
24 ScaleDepth-N 0.0740.2670.0320.9570.9940.999 – Paper Code 2024 linked, not harvested report
25 DepthGen 0.0740.3140.0320.9460.987 0.996 ✓ Paper – 2023 no code linked report
26 ZoeD-M12-N 0.0750.2700.0320.9550.9950.999 ✓ Paper Code 2023 10 of 16 ran · 6 unverified report
27 AiT-P(SwinV2-L) 0.0760.2750.0330.9540.9940.999 – Paper Code 2023 linked, not harvested report
28 Gaming for Depth (GfD) 0.0800.3640.0330.9310.9860.996 ✓ Paper – 2023 no code linked report
29 SwinV2-L 1K-MIM 0.0830.2870.0350.9490.9940.999 – Paper Code 2022 4 of 5 ran · 1 unverified report
30 Metric3D (ConvNeXt-Large, Zero-shot testing) 0.0830.3100.0350.9440.9860.995 ✓ Paper Code 2023 linked, not harvested report
31 VA-DepthNet(SwinV1-L) 0.0860.3040.0370.9370.9920.999 – Paper Code 2023 3 of 5 ran · 2 unverified report
32 iDisc 0.0860.9930.999 – Paper Code 2023 5 of 6 ran · 1 unverified report
33 MIM-Swin-V2 0.08640.30460.03650.93610.99160.9981 – Paper – 2023 no code linked report
34 NDDepth 0.0870.3110.0380.9360.9910.998 – Paper Code 2023 linked, not harvested report
35 IEBins 0.0870.3140.0380.9360.9920.998 – Paper Code 2023 8 of 13 ran · 5 unverified report
36 URCDC-Depth 0.0880.3160.0380.9330.9920.998 – Paper Code 2023 5 of 11 ran · 6 unverified report
37 OrdinalEntropy 0.0890.3210.0390.932 – Paper Code 2023 4 of 5 ran · 1 unverified report
38 PixelFormer 0.0900.3220.0390.9290.9910.998 – Paper Code 2022 5 of 11 ran · 6 unverified report
39 LeReS 0.090.916 ✓ Paper Code 2020 linked, not harvested report
40 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) 0.09070.2790.03710.94970.9960.9994 ✓ Paper Code 2023 21 of 46 ran · 25 unverified report
41 DDP (step3) 0.0940.3290.0400.9210.9900.998 – Paper Code 2023 linked, not harvested report
42 BinsFormer 0.0940.3300.0400.9250.9890.997 – Paper Code 2022 1 of 5 ran · 4 unverified report
43 NeWCRFs 0.0950.3340.0410.9220.9920.998 – Paper Code 2022 5 of 10 ran · 5 unverified report
44 D-Net 0.0950.3540.0410.9190.9880.997 – Paper Code 2021 linked, not harvested report
45 DepthFormer 0.0960.3390.0410.9210.9890.998 – Paper Code 2022 linked, not harvested report
46 GLPDepth 0.0980.3440.0420.9150.9880.997 – Paper Code 2022 1 of 2 ran · 1 unverified report
47 LocalBins 0.0980.3510.0420.910.9860.997 – Paper Code 2022 11 of 14 ran · 3 unverified report
48 Depth-Map-Decomposition-HRWSI 0.0980.3550.0420.9130.9870.998 ✓ Paper Code 2022 4 of 5 ran · 1 unverified report
49 Depthformer 0.1000.3450.0420.9130.9880.997 – Paper Code 2022 1 of 2 ran · 1 unverified report
50 Depth-Map-Decomposition 0.1000.3620.0430.9070.9860.997 – Paper Code 2022 4 of 5 ran · 1 unverified report
51 IronDepth 0.1010.3520.0430.9100.9850.997 – Paper Code 2022 linked, not harvested report
52 AdaBins 0.1030.3640.0440.9030.9840.997 – Paper Code 2020 linked, not harvested report
53 P3Depth 0.1040.3560.0430.8980.9810.996 – Paper Code 2022 linked, not harvested report
54 CutDepth 0.1040.3750.0440.8990.9850.997 ✓ Paper Code 2021 linked, not harvested report
55 LapDepth 0.1050.3840.0450.8950.9830.996 – Paper Code 2021 linked, not harvested report
56 DPT-Hybrid 0.1100.3570.0450.9040.9880.994 ✓ Paper Code 2021 54 of 116 ran · 62 unverified report
57 VNL 0.1110.4160.0480.8750.9760.989 – Paper Code 2019 linked, not harvested report
58 Focal-WNet 0.1160.3980.0480.8750.9800.995 – Paper Code 2022 linked, not harvested report
59 SC-DepthV2 0.1380.5320.0590.8200.956 – Paper Code 2020 linked, not harvested report
60 NVS-MonoDepth 0.331 – Paper – 2021 no code linked report
61 BTS 0.3920.995 – Paper Code 2019 5 of 15 ran · 10 unverified report
62 DSN 0.429 ✓ Paper – 2020 no code linked report
63 DenseDepth 0.465 – Paper Code 2018 5 of 23 ran · 18 unverified report
64 ACAN 0.496 – Paper Code 2019 0 of 12 ran · 12 unverified report
65 SharpNet 0.496 – Paper Code 2019 linked, not harvested report
66 PAP-Depth 0.497 – Paper – 2019 no code linked report
67 SDC-Depth 0.497 – Paper – 2020 no code linked report
68 DORN 0.509 – Paper Code 2018 0 of 3 ran · 3 unverified report
69 SARPN 0.514 – Paper Code 2019 linked, not harvested report
70 InvPT 0.5183 – Paper Code 2022 2 of 4 ran · 2 unverified report
71 FastDenseNas-arch0 0.523 – Paper Code 2018 linked, not harvested report
72 FastDenseNas-arch2 0.525 – Paper Code 2018 linked, not harvested report
73 FastDenseNas-arch1 0.526 – Paper Code 2018 linked, not harvested report
74 SENet-154 0.530 – Paper Code 2018 linked, not harvested report
75 ProbMonoDepth 0.536 – Paper Code 2019 linked, not harvested report
76 RelativeDepth 0.538 – Paper – 2019 no code linked report
77 PGT (Swin-S) 0.5468 – Paper Code 2023 linked, not harvested report
78 Index Network 0.565 – Paper Code 2019 linked, not harvested report
79 Multi-Task Light-Weight-RefineNet 0.565 – Paper Code 2018 6 of 6 ran · 0 unverified report
80 DeepLabV3+ (F10) 0.575 – Paper Code 2020 linked, not harvested report
81 Xu et al. 0.586 – Paper Code 2017 linked, not harvested report
82 PGT (Swin-T) 0.59 – Paper Code 2023 linked, not harvested report
83 SOM 0.604 – Paper – 2019 no code linked report
84 Li et al. 0.635 – Paper – 2016 no code linked report
85 Eigen et al. 0.641 – Paper Code 2014 0 of 3 ran · 3 unverified report

All 85 rows shown. 85 link to a paper page on this site; 22 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). 37 rows have a graph line, from 35 distinct papers; 34 rows (32 papers) have at least one sample that ran. Counting each paper once: Syntology ran 254 of 474 samples; 220 unverified. Separately, 155 of those 474 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