TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Computer Vision/Depth Estimation

Depth Estimation

173 benchmarks2454 papers

Depth Estimation is the task of measuring the distance of each pixel relative to the camera. Depth is extracted from either monocular (single) or stereo (multiple views of a scene) images. Traditional methods use multi-view geometry to find the relationship between the images. Newer methods can directly estimate depth by minimizing the regression loss, or by learning to generate a novel view from a sequence. The most popular benchmarks are KITTI and NYUv2. Models are typically evaluated according to a RMS metric.

<span class="description-source">Source: DIODE: A Dense Indoor and Outdoor DEpth Dataset </span>

Benchmarks

Depth Estimation on KITTI Eigen split

absolute relative errorDelta < 1.25RMSEDelta < 1.25^2Delta < 1.25^3RMSE logSq RelNumber of parameters (M)Square relative error (SqRel)

Depth Estimation on NYU-Depth V2

RMSEabsolute relative errorDelta < 1.25Delta < 1.25^2Delta < 1.25^3log 10RMSmAP

Depth Estimation on KITTI Eigen split unsupervised

absolute relative errorRMSESq RelRMSE logDelta < 1.25Delta < 1.25^2Delta < 1.25^3Test framesResolutionMono

Depth Estimation on Stanford2D3D Panoramic

RMSEabsolute relative error

Depth Estimation on ETH3D

Delta < 1.25absolute relative error

Depth Estimation on KITTI Depth Completion Validation

RMSE

Depth Estimation on NYU-Depth V2 self-supervised

Root mean square error (RMSE)Absolute relative error (AbsRel)delta_1delta_2delta_3

Depth Estimation on SiW (Protocol 3)

ACER

Depth Estimation on Make3D

RMSEAbs RelSq Rel

Depth Estimation on Mid-Air Dataset

Abs RelRMSE logRMSESQ RelAuSE on Abs RelAuSE on RMSE log

Depth Estimation on CelebA-Spoof-Enroll5

AUC

Depth Estimation on Helvipad

Depth-MAEDepth-RMSEDepth-MAREDepth-LRCEDisp-MAEDisp-RMSEDisp-MAREDisp-LRCE

Depth Estimation on SiW-Enroll5

AUC

Depth Estimation on DDAD

RMSEabsolute relative errorRMSE logSq RelDelta < 1.25

Depth Estimation on IBims-1

δ1.25RMSEORDD3Rabsolute relative error

Depth Estimation on KITTI2015

three pixel error three pixel errorD1-all AllD1-all Noc

Depth Estimation on OULU-NPU

ACERHTER

Depth Estimation on Replay-Attack

EERHTER

Depth Estimation on SCARED-C

mDERS

Depth Estimation on Spring

1px total

Depth Estimation on Cityscapes

RMSERMSE logAbsolute relative error (AbsRel)Square relative error (SqRel)

Depth Estimation on DCM

Abs RelRMSERMSE logSq Rel

Depth Estimation on DIODE

Delta < 1.25^3Delta < 1.25Delta < 1.25^2

Depth Estimation on Relative Human

PCDRPCDR-BabyPCDR-KidPCDR-TeenPCDR-AdultmPCDK

Depth Estimation on SUN-RGBD

Delta < 1.25RMSEabsolute relative errorDelta < 1.25^2Delta < 1.25^3log 10

Depth Estimation on ScanNetV2

absolute relative errorDelta < 1.25

Depth Estimation on VA (Virtual Apartment)

Root mean square error (RMSE)Log root mean square error (RMSE_log)Mean average error (MAE) Absolute relative error (AbsRel)

Depth Estimation on eBDtheque

Abs RelRMSERMSE logSq Rel

Depth Estimation on sceneflow

Average End-Point ErrorEPE

Depth Estimation on CASIA-MFSD

EERHTER

Depth Estimation on Cityscapes test

RMSE

Depth Estimation on KITTI

absolute relative error

Depth Estimation on KITTI 2015

Absolute relative error (AbsRel)Sq RelD1-all AllD1-all NocRMSE

Depth Estimation on Mars DTM Estimation

Delta < 1.25Delta < 1.25^2Delta < 1.25^3Average PSNRmean absolute errorRMSE

Depth Estimation on Middlebury 2014

ORD D3RRMSEδ1.25

Depth Estimation on ScanNet

RMSEabsolute relative error

Depth Estimation on 4D Light Field Dataset

BadPix(0.01)BadPix(0.03)BadPix(0.07)MSE

Depth Estimation on Cityscapes 3D

RMSE

Depth Estimation on DIML Outdoor

Delta < 1.25RMSEabsolute relative error

Depth Estimation on DIODE Indoor

Delta < 1.25RMSEabsolute relative error

Depth Estimation on DIODE Outdoor

Delta < 1.25RMSEabsolute relative error

Depth Estimation on Hypersim

Delta < 1.25RMSEabsolute relative error

Depth Estimation on KITTI Object Tracking Evaluation 2012

Abs Rel

Depth Estimation on KITTI2012

three pixel error

Depth Estimation on MLFP

HTER

Depth Estimation on MSU-MFSD

Equal Error RateHTER

Depth Estimation on Matterport3D

Abs RelDelta < 1.25Delta < 1.25^2Delta < 1.25^3RMSEabsolute errorabsolute relative error

Depth Estimation on Taskonomy

L1 error

Depth Estimation on TransCG

Delta < 1.25MAERELRMSEdelta < 1.05delta < 1.10

Depth Estimation on UASOL

RMSE

Depth Estimation on Virtual KITTI 2

Delta < 1.25RMSEabsolute relative error