</> Code with Papers by Syntology

Datasets › KITTI

KITTI

Introduced in Are we ready for autonomous driving? The KITTI vision benchmark suite1 Jan 2012 archive 2025-07-28

KITTI (Karlsruhe Institute of Technology and Toyota Technological Institute) is one of the most popular datasets for use in mobile robotics and autonomous driving. It consists of hours of traffic scenarios recorded with a variety of sensor modalities, including high-resolution RGB, grayscale stereo cameras, and a 3D laser scanner. Despite its popularity, the dataset itself does not contain ground truth for semantic segmentation. However, various researchers have manually annotated parts of the dataset to fit their necessities. Álvarez et al. generated ground truth for 323 images from the road detection challenge with three classes: road, vertical, and sky. Zhang et al. annotated 252 (140 for training and 112 for testing) acquisitions – RGB and Velodyne scans – from the tracking challenge for ten object categories: building, sky, road, vegetation, sidewalk, car, pedestrian, cyclist, sign/pole, and fence. Ros et al. labeled 170 training images and 46 testing images (from the visual odometry challenge) with 11 classes: building, tree, sky, car, sign, road, pedestrian, fence, pole, sidewalk, and bicyclist.

Source: A Review on Deep Learning Techniques Applied to Semantic Segmentation Image Source: http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 137 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Monocular Depth Estimation KITTI Eigen split SPIDepth absolute relative error 0.029 SPIdepth: Strengthened Pose Information for... Lavreniuk/SPIdepth 79 Compare
Monocular Depth Estimation KITTI Eigen split unsupervised SPIdepth absolute relative error 0.071 SPIdepth: Strengthened Pose Information for... Lavreniuk/SPIdepth 55 Compare
Multiple Object Tracking KITTI Test (Online Methods) RobMOT (Dynamic) HOTA 81.80 Towards Accurate State Estimation: Kalman Filter... — 34 Compare
Monocular 3D Object Detection KITTI Cars Moderate CIE AP Medium 20.95 Consistency of Implicit and Explicit Features Matters... — 29 Compare
3D Object Detection KITTI Cars Easy TRTConv AP 91.90 % — — 26 Compare
3D Object Detection KITTI Cars Hard TRTConv AP 80.38 % — — 25 Compare
Optical Flow Estimation KITTI 2015 (train) MEMFOF F1-all 9.93 MEMFOF: High-Resolution Training for Memory-Efficient... msu-video-group/memfof 19 Compare
Vehicle Pose Estimation KITTI Cars Hard Ego-Net (Monocular RGB only) Average Orientation Similarity 80.96 Exploring intermediate representation for monocular... Nicholasli1995/EgoNet 19 Compare
Optical Flow Estimation KITTI 2015 MEMFOF Fl-all 2.94 MEMFOF: High-Resolution Training for Memory-Efficient... msu-video-group/memfof 18 Compare
Point Cloud Registration KITTI (trained on 3DMatch) GeDi Success Rate 98.92 Learning general and distinctive 3D local deep... fabiopoiesi/gedi 14 Compare
3D Object Detection KITTI Cyclists Moderate 3D-FCT AP 75.86% 3D-FCT: Simultaneous 3D Object Detection and Tracking... — 13 Compare
3D Object Detection From Stereo Images KITTI Cars Moderate DSGN++ AP75 67.37 DSGN++: Exploiting Visual-Spatial Relation for... chenyilun95/dsgn2 12 Compare
3D Object Detection KITTI Cyclists Easy 3D-FCT AP 89.15% 3D-FCT: Simultaneous 3D Object Detection and Tracking... — 12 Compare
3D Object Detection KITTI Cyclists Hard SA-Det3D AP 61.33% SA-Det3D: Self-Attention Based Context-Aware 3D Object Detection AutoVision-cloud/SA-Det3D 12 Compare
3D Object Detection KITTI Pedestrians Moderate 3D-FCT AP 58.4% 3D-FCT: Simultaneous 3D Object Detection and Tracking... — 12 Compare
Optical Flow Estimation KITTI 2012 CroCo-Flow Average End-Point Error 0.8 CroCo v2: Improved Cross-view Completion Pre-training... naver/croco 12 Compare
3D Object Detection KITTI Cars Easy val SA-SSD+EBM AP 95.45 Accurate 3D Object Detection using Energy-Based Models fregu856/ebms_3dod 11 Compare
3D Object Detection KITTI Cars Moderate val SA-SSD+EBM AP 86.83 Accurate 3D Object Detection using Energy-Based Models fregu856/ebms_3dod 11 Compare
Point Cloud Registration KITTI (FCGF setting) GeoTransformer Recall (0.6m, 5 degrees) 99.5 Geometric Transformer for Fast and Robust Point Cloud... qinzheng93/geotransformer +1 11 Compare
3D Object Detection KITTI Cars Hard val M3DeTR AP 82.85 M3DeTR: Multi-representation, Multi-scale,... rayguan97/M3DeTR 10 Compare
3D Object Detection KITTI Pedestrians Easy IPOD AP 56.92% IPOD: Intensive Point-based Object Detector for Point Cloud — 9 Compare
3D Object Detection KITTI Pedestrians Hard SVGA-Net AP 44.56% SVGA-Net: Sparse Voxel-Graph Attention Network for 3D... — 9 Compare
Birds Eye View Object Detection KITTI Cars Moderate SE-SSD AP 91.84% SE-SSD: Self-Ensembling Single-Stage Object Detector... Vegeta2020/SE-SSD 9 Compare
Birds Eye View Object Detection KITTI Cars Easy SE-SSD AP 95.68% SE-SSD: Self-Ensembling Single-Stage Object Detector... Vegeta2020/SE-SSD 9 Compare
Stereo Image Super-Resolution KITTI2012 - 4x upscaling SwinFIRSSR PSNR 27.16 SwinFIR: Revisiting the SwinIR with Fast Fourier... Zdafeng/SwinFIR +1 9 Compare
Stereo Image Super-Resolution KITTI2015 - 2x upscaling ASteISR PSNR 31.48 ASteISR: Adapting Single Image Super-resolution... fzuzyb/ASteISR 9 Compare
Stereo Image Super-Resolution KITTI2015 - 4x upscaling NAFSSR-L PSNR 26.96 NAFSSR: Stereo Image Super-Resolution Using NAFNet megvii-research/NAFNet +4 9 Compare
Birds Eye View Object Detection KITTI Cars Hard STD AP 86.89 STD: Sparse-to-Dense 3D Object Detector for Point Cloud — 8 Compare
Monocular 3D Object Detection KITTI Cars Hard CIE AP Hard 17.83 Consistency of Implicit and Explicit Features Matters... — 7 Compare
Semantic Segmentation KITTI Semantic Segmentation RPVNet [xu2021rpvnet] Mean IoU (class) 80.7 Spherical Transformer for LiDAR-based 3D Recognition dvlab-research/sphereformer +1 7 Compare
Stereo Depth Estimation KITTI2015 AnyNet three pixel error 6.2 Anytime Stereo Image Depth Estimation on Mobile Devices mileyan/AnyNet +2 7 Compare
3D Object Detection From Stereo Images KITTI Pedestrians Moderate DSGN++ AP50 32.74 DSGN++: Exploiting Visual-Spatial Relation for... chenyilun95/dsgn2 6 Compare
Birds Eye View Object Detection KITTI Cyclists Moderate PV-RCNN AP 68.89% PV-RCNN: Point-Voxel Feature Set Abstraction for 3D... open-mmlab/OpenPCDet +11 6 Compare
Birds Eye View Object Detection KITTI Pedestrians Moderate Frustrum-PointPillars AP 52.23 % Frustum-PointPillars: A Multi-Stage Approach for 3D... anshulpaigwar/Frustum-Pointpillars +1 6 Compare
Point Cloud Registration KITTI GeDi Success Rate 99.82 Learning general and distinctive 3D local deep... fabiopoiesi/gedi 6 Compare
3D Object Detection From Stereo Images KITTI Cyclists Moderate DSGN++ AP50 43.90 DSGN++: Exploiting Visual-Spatial Relation for... chenyilun95/dsgn2 5 Compare
Monocular 3D Object Detection KITTI Cars Easy CIE AP Easy 31.55 Consistency of Implicit and Explicit Features Matters... — 5 Compare
Object Detection KITTI Cars Easy Patches AP 87.87 Patch Refinement -- Localized 3D Object Detection — 5 Compare
Object Detection KITTI Cars Hard Patches AP 68.91 Patch Refinement -- Localized 3D Object Detection — 5 Compare
Stereo Image Super-Resolution KITTI2012 - 2x upscaling SwinFIRSSR PSNR 31.79 SwinFIR: Revisiting the SwinIR with Fast Fourier... Zdafeng/SwinFIR +1 5 Compare
Stereo Image Super-Resolution KITTI2012 - 2x upscaling ASteISR PSNR 31.86 ASteISR: Adapting Single Image Super-resolution... fzuzyb/ASteISR 5 Compare
3D Object Detection KITTI Cyclist Easy val M3DeTR AP 89.13 M3DeTR: Multi-representation, Multi-scale,... rayguan97/M3DeTR 4 Compare
3D Object Detection KITTI Cyclist Hard val M3DeTR AP 68.29 M3DeTR: Multi-representation, Multi-scale,... rayguan97/M3DeTR 4 Compare
3D Object Detection KITTI Cyclist Moderate val M3DeTR AP 71.70 M3DeTR: Multi-representation, Multi-scale,... rayguan97/M3DeTR 4 Compare
3D Object Detection KITTI Pedestrian Moderate val PVCNN AP 64.71 Point-Voxel CNN for Efficient 3D Deep Learning isl-org/Open3D-ML +3 4 Compare
3D Object Detection KITTI Pedestrian Easy val PVCNN AP 73.2 Point-Voxel CNN for Efficient 3D Deep Learning isl-org/Open3D-ML +3 4 Compare
3D Object Detection KITTI Pedestrian Hard val PVCNN AP 56.78 Point-Voxel CNN for Efficient 3D Deep Learning isl-org/Open3D-ML +3 4 Compare
Monocular 3D Object Detection KITTI Pedestrian Hard DD3D AP Hard 8.05 Is Pseudo-Lidar needed for Monocular 3D Object detection? tri-ml/dd3d +1 4 Compare
Object Detection KITTI Cars Moderate Patches AP 77.16 Patch Refinement -- Localized 3D Object Detection — 4 Compare
Optical Flow Estimation KITTI 2015 unsupervised MDFlow Fl-all 8.91 MDFlow: Unsupervised Optical Flow Learning by Reliable... ltkong218/mdflow 4 Compare
Panoptic Segmentation KITTI Panoptic Segmentation EfficientPS PQ 43.7 EfficientPS: Efficient Panoptic Segmentation DeepSceneSeg/EfficientPS +1 4 Compare
Scene Flow Estimation KITTI 2015 Scene Flow Training EPC++ D1-all 23.84 Every Pixel Counts ++: Joint Learning of Geometry and... chenxuluo/EPC 4 Compare
Scene Flow Estimation KITTI 2015 Scene Flow Test CamLiRAFT SF-all 4.26 Learning Optical Flow and Scene Flow with Bidirectional... mcg-nju/camliflow 4 Compare
Unsupervised Panoptic Segmentation KITTI CUPS (54 pseudo-classes) PQ 28.5 Scene-Centric Unsupervised Panoptic Segmentation visinf/cups 4 Compare
Birds Eye View Object Detection KITTI Cars Moderate val VoxelNet AP 84.81 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 3 Compare
Image Dehazing KITTI FFA-Net PSNR 27.45 FFA-Net: Feature Fusion Attention Network for Single... zhilin007/FFA-Net +3 3 Compare
Monocular 3D Object Detection KITTI Pedestrian Easy CMKD AP Easy 13.94 Cross-Modality Knowledge Distillation Network for... Cc-Hy/CMKD 3 Compare
Monocular 3D Object Detection KITTI Pedestrian Moderate DD3D AP Medium 9.30 Is Pseudo-Lidar needed for Monocular 3D Object detection? tri-ml/dd3d +1 3 Compare
Object Localization KITTI Pedestrians Moderate Frustrum-PointPillars AP 52.23 % Frustum-PointPillars: A Multi-Stage Approach for 3D... anshulpaigwar/Frustum-Pointpillars +1 3 Compare
Object Localization KITTI Pedestrians Hard Frustrum-PointPillars AP 48.30 % Frustum-PointPillars: A Multi-Stage Approach for 3D... anshulpaigwar/Frustum-Pointpillars +1 3 Compare
Video Prediction KITTI DMVFN LPIPS 0.1074 A Dynamic Multi-Scale Voxel Flow Network for Video Prediction megvii-research/CVPR2023-DMVFN 3 Compare
Birds Eye View Object Detection KITTI Pedestrians Easy STD AP 60.99 STD: Sparse-to-Dense 3D Object Detector for Point Cloud — 2 Compare
Birds Eye View Object Detection KITTI Pedestrians Hard Frustrum-PointPillars AP 48.30 % Frustum-PointPillars: A Multi-Stage Approach for 3D... anshulpaigwar/Frustum-Pointpillars +1 2 Compare
Birds Eye View Object Detection KITTI Cyclists Easy PV-RCNN AP 82.49 PV-RCNN: Point-Voxel Feature Set Abstraction for 3D... open-mmlab/OpenPCDet +11 2 Compare
Birds Eye View Object Detection KITTI Cyclists Hard PV-RCNN AP 62.41 PV-RCNN: Point-Voxel Feature Set Abstraction for 3D... open-mmlab/OpenPCDet +11 2 Compare
Birds Eye View Object Detection KITTI Cars Easy val VoxelNet AP 89.6 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 2 Compare
Birds Eye View Object Detection KITTI Cars Hard val VoxelNet AP 78.57 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 2 Compare
Dense Pixel Correspondence Estimation KITTI 2012 COTR Average End-Point Error 1.28 COTR: Correspondence Transformer for Matching Across Images ubc-vision/COTR 2 Compare
Dense Pixel Correspondence Estimation KITTI 2015 COTR Average End-Point Error 2.26 COTR: Correspondence Transformer for Matching Across Images ubc-vision/COTR 2 Compare
Depth Estimation KITTI 2015 H-Net (Ours) Full Eigen Absolute relative error (AbsRel) 0.076 H-Net: Unsupervised Attention-based Stereo Depth... — 2 Compare
Monocular 3D Object Detection KITTI Cyclist Easy CMKD AP Easy 12.52 Cross-Modality Knowledge Distillation Network for... Cc-Hy/CMKD 2 Compare
Monocular 3D Object Detection KITTI Cyclist Moderate CMKD AP Medium 6.67 Cross-Modality Knowledge Distillation Network for... Cc-Hy/CMKD 2 Compare
Monocular 3D Object Detection KITTI Cyclist Hard CMKD AP Hard 6.34 Cross-Modality Knowledge Distillation Network for... Cc-Hy/CMKD 2 Compare
Monocular Cross-View Road Scene Parsing(Vehicle) KITTI2012 DCTNet mAP 58.89% A Dual-Cycled Cross-View Transformer Network for Unified... AutoCompSysLab/DCTNet 2 Compare
Monocular Cross-View Road Scene Parsing(Road) Kitti Odometry DCTNet mAP 88.28% A Dual-Cycled Cross-View Transformer Network for Unified... AutoCompSysLab/DCTNet 2 Compare
Monocular Cross-View Road Scene Parsing(Road) Kitti Raw DCTNet mAP 86.56% A Dual-Cycled Cross-View Transformer Network for Unified... AutoCompSysLab/DCTNet 2 Compare
Monocular Depth Estimation KITTI MonoViT absolute relative error 0.093 MonoViT: Self-Supervised Monocular Depth Estimation with... zxcqlf/monovit 2 Compare
Multiple Object Tracking KITTI Test (Offline Methods) MCTrack HOTA 82.75 MCTrack: A Unified 3D Multi-Object Tracking Framework... megvii-research/mctrack 2 Compare
Object Localization KITTI Cars Easy VoxelNet AP 89.35% VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 2 Compare
Object Localization KITTI Cars Hard VoxelNet AP 77.39% VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 2 Compare
Object Localization KITTI Cars Moderate Frustum PointNets AP 84.0% Frustum PointNets for 3D Object Detection from RGB-D Data charlesq34/pointnet +67 2 Compare
Object Localization KITTI Cyclists Moderate Frustum PointNets AP 61.96% Frustum PointNets for 3D Object Detection from RGB-D Data charlesq34/pointnet +67 2 Compare
Object Localization KITTI Cyclists Easy Frustum PointNets AP 75.38% Frustum PointNets for 3D Object Detection from RGB-D Data charlesq34/pointnet +67 2 Compare
Object Localization KITTI Cyclists Hard Frustum PointNets AP 54.68% Frustum PointNets for 3D Object Detection from RGB-D Data charlesq34/pointnet +67 2 Compare
Object Localization KITTI Pedestrians Easy Frustum PointNets AP 58.09% Frustum PointNets for 3D Object Detection from RGB-D Data charlesq34/pointnet +67 2 Compare
Object Tracking KITTI M2-Track mean precision 83.4 Beyond 3D Siamese Tracking: A Motion-Centric Paradigm... ghostish/open3dsot 2 Compare
Optical Flow Estimation KITTI 2012 unsupervised UpFlow Average End-Point Error 1.4 UPFlow: Upsampling Pyramid for Unsupervised Optical Flow Learning twhui/LiteFlowNet3 +1 2 Compare
Stereo Depth Estimation KITTI 2015 MoCha-Stereo D1-all All 1.53 MoCha-Stereo: Motif Channel Attention Network for Stereo Matching zyangchen/mocha-stereo 2 Compare
Stereo Disparity Estimation KITTI 2015 MoCha-Stereo D1-all 1.53 MoCha-Stereo: Motif Channel Attention Network for Stereo Matching zyangchen/mocha-stereo 2 Compare
3D Object Detection KITTI Cyclists Moderate val Deformable PV-RCNN AP 73.46 Deformable PV-RCNN: Improving 3D Object Detection with... AutoVision-cloud/DeformablePVRCNN +1 1 Compare
3D Object Detection KITTI Pedestrian Moderate PiFeNet Average Precision 0.4671 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
3D Object Detection KITTI Pedestrian PiFeNet mAP 0.486 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
3D Object Detection KITTI Pedestrian Easy PiFeNet Average Precision 0.5639 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
3D Object Detection KITTI Pedestrian Hard PiFeNet Average Precision 0.4271 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
3D Object Detection KITTI Pedestrians Moderate val Deformable PV-RCNN AP 58.33 Deformable PV-RCNN: Improving 3D Object Detection with... AutoVision-cloud/DeformablePVRCNN +1 1 Compare
Birds Eye View Object Detection KITTI Pedestrian Easy PiFeNet Average Precision 0.6325 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
Birds Eye View Object Detection KITTI Pedestrian Moderate PiFeNet Average Precision 0.5392 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
Birds Eye View Object Detection KITTI Pedestrian Hard PiFeNet Average Precision 0.5053 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
Birds Eye View Object Detection KITTI Pedestrian PiFeNet mAP 0.559 Accurate and Real-time 3D Pedestrian Detection Using an... ldtho/pifenet 1 Compare
Birds Eye View Object Detection KITTI Pedestrian Easy val VoxelNet AP 65.95 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 1 Compare
Birds Eye View Object Detection KITTI Pedestrian Moderate val VoxelNet AP 61.05 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 1 Compare
Birds Eye View Object Detection KITTI Pedestrian Hard val VoxelNet AP 56.98 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 1 Compare
Birds Eye View Object Detection KITTI Cyclist Easy val VoxelNet AP 74.41 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 1 Compare
Birds Eye View Object Detection KITTI Cyclist Moderate val VoxelNet AP 52.18 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 1 Compare
Birds Eye View Object Detection KITTI Cyclist Hard val VoxelNet AP 50.49 VoxelNet: End-to-End Learning for Point Cloud Based 3D... qianguih/voxelnet +43 1 Compare
Depth Completion KITTI FusionDepth RMSE 1193.92 Advancing Self-supervised Monocular Depth Learning with... AutoAILab/FusionDepth +1 1 Compare
Depth Estimation KITTI Eigen split LightDepth Number of parameters (M) 42.6 LightDepth: A Resource Efficient Depth Estimation... fatemehkarimii/lightdepth 1 Compare
Egocentric Pose Estimation Kitti Odometry pc4consistentdepth Absolute Trajectory Error [m] 0.014 Pose Constraints for Consistent Self-supervised... zshn25/pc4consistentdepth 1 Compare
Horizon Line Estimation KITTI Horizon ConvLSTM (Huber Loss, naive residual path) ATV 4.984 Temporally Consistent Horizon Lines fkluger/tchl 1 Compare
Image Clustering KITTI TURTLE (CLIP + DINOv2) Accuracy 39.4 Let Go of Your Labels with Unsupervised Transfer mlbio-epfl/turtle 1 Compare
Image Super-Resolution KITTI 2012 - 2x upscaling PASSRnet PSNR 30.65 Learning Parallax Attention for Stereo Image Super-Resolution LongguangWang/PASSRnet 1 Compare
Image Super-Resolution KITTI 2012 - 4x upscaling PASSRnet PSNR 26.26 Learning Parallax Attention for Stereo Image Super-Resolution LongguangWang/PASSRnet 1 Compare
Image Super-Resolution KITTI 2015 - 2x upscaling PASSRnet PSNR 29.78 Learning Parallax Attention for Stereo Image Super-Resolution LongguangWang/PASSRnet 1 Compare
Image Super-Resolution KITTI 2015 - 4x upscaling PASSRnet PSNR 25.43 Learning Parallax Attention for Stereo Image Super-Resolution LongguangWang/PASSRnet 1 Compare
Image-to-Image Translation KITTI Object Tracking Evaluation 2012 SRNet Average PSNR 21.12 Editing Text in the Wild youdao-ai/SRNet +1 1 Compare
Image to Point Cloud Registration KITTI CorrI2P RRE 2.07 CorrI2P: Deep Image-to-Point Cloud Registration via... rsy6318/CorrI2P 1 Compare
Knowledge Distillation KITTI TIE-KD (T: Adabins S: MobileNetV2) RMSE 2.4315 TIE-KD: Teacher-Independent and Explainable Knowledge... hpc-lab-koreatech/tie-kd 1 Compare
Monocular 3D Object Detection KITTI Pedestrians Moderate val CubifAE-3D AP Medium 5.43 CubifAE-3D: Monocular Camera Space Cubification for... — 1 Compare
Monocular Depth Estimation KITTI Object Tracking Evaluation 2012 PackNet-SfM Abs Rel 0.071 3D Packing for Self-Supervised Monocular Depth Estimation TRI-ML/packnet-sfm +3 1 Compare
Novel View Synthesis KITTI READ Average PSNR 23.28 READ: Large-Scale Neural Scene Rendering for Autonomous Driving JOP-Lee/READ-Large-Scale-Neural-Scene-Rendering-for-Autonomous-Driving 1 Compare
Novel View Synthesis KITTI Novel View Synthesis Multi-view to Novel View SSIM 0.626 Multi-view to Novel view: Synthesizing Novel Views with... shaohua0116/Multiview2Novelview 1 Compare
Object Detection KITTI Cyclists Easy Vote3Deep AP 79.92 Vote3Deep: Fast Object Detection in 3D Point Clouds... — 1 Compare
Object Detection KITTI Cyclists Hard Vote3Deep AP 62.98 Vote3Deep: Fast Object Detection in 3D Point Clouds... — 1 Compare
Object Detection KITTI Cyclists Moderate Vote3Deep AP 67.88 Vote3Deep: Fast Object Detection in 3D Point Clouds... — 1 Compare
Object Detection KITTI Pedestrians Moderate Vote3Deep AP 55.37 Vote3Deep: Fast Object Detection in 3D Point Clouds... — 1 Compare
Object Detection KITTI Pedestrians Easy Vote3Deep AP 68.39 Vote3Deep: Fast Object Detection in 3D Point Clouds... — 1 Compare
Object Detection KITTI Pedestrians Hard Vote3Deep AP 52.59 Vote3Deep: Fast Object Detection in 3D Point Clouds... — 1 Compare
Object Localization KITTI Pedestrian Easy Frustrum-PointPillars AP 60.98 % Frustum-PointPillars: A Multi-Stage Approach for 3D... anshulpaigwar/Frustum-Pointpillars +1 1 Compare
Pose Estimation KITTI 2015 GeoNet Average End-Point Error 10.81 GeoNet: Unsupervised Learning of Dense Depth, Optical... yzcjtr/GeoNet +2 1 Compare
Real-time Instance Segmentation KITTI CenterPoly AP 8.73 CenterPoly: real-time instance segmentation using... hu64/centerpoly 1 Compare
Scene Generation KITTI GaussianCity FID 29.5 GaussianCity: Generative Gaussian Splatting for... hzxie/GaussianCity 1 Compare
Stereo Depth Estimation KITTI2012 AnyNet three pixel error 6.1 Anytime Stereo Image Depth Estimation on Mobile Devices mileyan/AnyNet +2 1 Compare
Stereo Image Super-Resolution KITTI2012 - 2x scaling LSSR PSNR 31.40 Learning Optimal Combination Patterns for Lightweight... — 1 Compare
Text-To-SQL 2D KITTI Cars Easy sdfa 0..5sec dafa Non-local Neural Networks facebookresearch/detectron +31 1 Compare
Transfer Learning KITTI Object Tracking Evaluation 2012 Physical Access EER 5.74 Audio Spoofing Verification using Deep Convolutional... rahul-t-p/ASVspoof-2019 1 Compare
Vehicle Pose Estimation KITTI Ego-Net Average Orientation Similarity 89.43 Exploring intermediate representation for monocular... Nicholasli1995/EgoNet 1 Compare
Visual Place Recognition KITTI None Average F1 0.932 SSC: Semantic Scan Context for Large-Scale Place Recognition lilin-hitcrt/SSC 1 Compare

Papers archive 2025-07-28

30 shown of 335 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 3,661. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
MEMFOF: High-Resolution Training for Memory-Efficient Multi-Frame Optical Flow Estimation 1 2 29 Jun 2025 ran 9 of 13 samples (4 unverified)
Towards Accurate State Estimation: Kalman Filter Incorporating Motion Dynamics for 3D Multi-Object Tracking 0 1 12 May 2025 not harvested
Scene-Centric Unsupervised Panoptic Segmentation 1 3 2 Apr 2025 ran 0 of 8 samples (8 unverified)
UniK3D: Universal Camera Monocular 3D Estimation 1 1 20 Mar 2025 not harvested
Jasmine: Harnessing Diffusion Prior for Self-supervised Depth Estimation 0 1 20 Mar 2025 not harvested
DPFlow: Adaptive Optical Flow Estimation with a Dual-Pyramid Framework 1 2 19 Mar 2025 not harvested
UniDepthV2: Universal Monocular Metric Depth Estimation Made Simpler 1 1 27 Feb 2025 ran 2 of 2 samples (0 unverified; 2 pointer-only for licence)
DepthMaster: Taming Diffusion Models for Monocular Depth Estimation 1 1 5 Jan 2025 not harvested
Learning Optimal Combination Patterns for Lightweight Stereo Image Super-Resolution 0 4 28 Oct 2024 not harvested
MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors 1 3 25 Oct 2024 not harvested
MCTrack: A Unified 3D Multi-Object Tracking Framework for Autonomous Driving 1 2 23 Sep 2024 not harvested
PrimeDepth: Efficient Monocular Depth Estimation with a Stable Diffusion Preimage 1 2 13 Sep 2024 ran 13 of 15 samples (2 unverified)
One Homography is All You Need: IMM-based Joint Homography and Multiple Object State Estimation 1 1 4 Sep 2024 not harvested
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training 1 7 26 Aug 2024 not harvested
ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion 1 2 12 Jul 2024 not harvested
ScaleDepth: Decomposing Metric Depth Estimation into Scale Prediction and Relative Depth Estimation 1 1 11 Jul 2024 not harvested
SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning 1 2 7 Jul 2024 not harvested
ASteISR: Adapting Single Image Super-resolution Pre-trained Model for Efficient Stereo Image Super-resolution 1 2 4 Jul 2024 not harvested
BiTrack: Bidirectional Offline 3D Multi-Object Tracking Using Camera-LiDAR Data 1 1 26 Jun 2024 not harvested
Learning Accurate and Enriched Features for Stereo Image Super-Resolution 1 4 23 Jun 2024 not harvested
Let Go of Your Labels with Unsupervised Transfer 1 1 11 Jun 2024 ran 3 of 4 samples (1 unverified; 4 pointer-only for licence)
GaussianCity: Generative Gaussian Splatting for Unbounded 3D City Generation 1 1 10 Jun 2024 ran 7 of 10 samples (3 unverified; 10 pointer-only for licence)
DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation 0 3 27 May 2024 not harvested
RobMOT: Robust 3D Multi-Object Tracking by Observational Noise and State Estimation Drift Mitigation on LiDAR PointCloud 1 1 19 May 2024 not harvested
RAPIDFlow: Recurrent Adaptable Pyramids with Iterative Decoding for Efficient Optical Flow Estimation 1 2 1 May 2024 not harvested
SPIdepth: Strengthened Pose Information for Self-supervised Monocular Depth Estimation 1 2 18 Apr 2024 ran 13 of 14 samples (1 unverified)
MoCha-Stereo: Motif Channel Attention Network for Stereo Matching 1 2 10 Apr 2024 ran 16 of 17 samples (1 unverified)
UniDepth: Universal Monocular Metric Depth Estimation 3 1 27 Mar 2024 ran 15 of 26 samples (11 unverified; 25 pointer-only for licence)
ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation 1 1 27 Mar 2024 ran 11 of 15 samples (4 unverified; 15 pointer-only for licence)
Addressing the generalization of 3D registration methods with a featureless baseline and an unbiased benchmark 1 2 23 Mar 2024 not harvested

The full list of 335 is in the JSON twin.

Dataset loaders archive 2025-07-28

pytorch.org/vision/stable/generated/torchvision.datasets.KittiFlow.htmlpytorch/visionpytorch
open-mmlab/mmdetection3dopen-mmlab/mmdetection3dpytorch
www.tensorflow.org/datasets/catalog/kittitensorflow/datasetstfjax

3 loaders as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

Object DetectionSemantic SegmentationPose EstimationObject Tracking3D Object DetectionImage Super-ResolutionImage-to-Image TranslationImage ClusteringDepth EstimationVisual Place RecognitionObject LocalizationPanoptic SegmentationPoint Cloud RegistrationMonocular Depth EstimationNovel View SynthesisVideo PredictionBirds Eye View Object DetectionTransfer LearningMonocular 3D Object DetectionImage DehazingMultiple Object TrackingText-To-SQLKnowledge DistillationOptical Flow EstimationStereo Image Super-ResolutionUnsupervised Object DetectionVisual OdometryDepth CompletionReal-time Instance SegmentationUnsupervised Panoptic SegmentationStereo Depth EstimationScene Flow EstimationScene GenerationDense Pixel Correspondence EstimationUnsupervised Monocular Depth EstimationHorizon Line EstimationEgocentric Pose EstimationVehicle Pose Estimation3D Object Detection From Stereo ImagesStereo Disparity EstimationMonocular Cross-View Road Scene Parsing(Road)Depth PredictionPrediction Of Occupancy Grid Maps3D Object TrackingMonocular Cross-View Road Scene Parsing(Vehicle)Image to Point Cloud Registration3D Single Object Tracking

License archive 2025-07-28

CC BY-NC-SA 3.0

Modalities archive 2025-07-28

ImagesLiDAR

Languages archive 2025-07-28

EnglishChinese

Variants archive 2025-07-28

  • KITTI Test (Offline Methods)
  • KITTI Test (Online Methods)
  • KITTI Cyclists Moderate val
  • KITTI (trained on 3DMatch)
  • KITTI 2015 (train)
  • KITTI2015 - 4x upscaling
  • KITTI2015 - 2x upscaling
  • KITTI2012 - 4x upscaling
  • KITTI2012 - 2x upscaling
  • KITTI2012 - 2x upscaling
  • KITTI2012 - 2x scaling
  • KITTI Pedestrian Moderate
  • KITTI Eigen Split Improved Ground Truth
  • KITTI Cyclist Moderate
  • KITTI Object Tracking Evaluation 2012
  • KITTI Pedestrian Easy
  • KITTI 2015 (train)
  • 2D KITTI Pedestrians Moderate
  • 2D KITTI Pedestrians Hard
  • 2D KITTI Pedestrians Easy
  • 2D KITTI Cyclists Moderate
  • 2D KITTI Cyclists Hard
  • 2D KITTI Cyclists Easy
  • Kitti Odometry
  • KITTI Cyclist Hard
  • KITTI Cyclist Easy
  • KITTI2012 Tracking
  • 2D KITTI Cars Hard
  • 2D KITTI Cars Easy
  • KITTI Stereo 2015
  • KITTI Stereo 2012
  • Kitti Raw
  • KITTI Pedestrian
  • KITTI Cars Simple
  • KITTI (FCGF setting)
  • 2D KITTI Cars Moderate
  • KITTI 2015 Scene Flow Test
  • KITTI 2015 Scene Flow Training
  • KITTI Novel View Synthesis
  • KITTI 2015 - unsupervised
  • KITTI 2012 - unsupervised
  • KITTI Pedestrians Moderate val
  • KITTI Pedestrian Hard
  • KITTI Panoptic Segmentation
  • KITTI Horizon
  • KITTI Tracking test
  • KITTI2015
  • KITTI 2015 unsupervised
  • KITTI 2015 - 4x upscaling
  • KITTI 2015 - 2x upscaling
  • KITTI 2015
  • KITTI2012
  • KITTI 2012 unsupervised
  • KITTI 2012 - 4x upscaling
  • KITTI 2012 - 2x upscaling
  • KITTI 2012
  • KITTI Semantic Segmentation
  • KITTI Pedestrians Moderate
  • KITTI Pedestrians Hard
  • KITTI Pedestrians Easy
  • KITTI Pedestrian Moderate val
  • KITTI Pedestrian Hard val
  • KITTI Pedestrian Easy val
  • KITTI Eigen split unsupervised
  • KITTI Eigen split
  • KITTI Cyclists Moderate
  • KITTI Cyclists Hard
  • KITTI Cyclists Easy
  • KITTI Cyclist Moderate val
  • KITTI Cyclist Hard val
  • KITTI Cyclist Easy val
  • KITTI Cars Moderate val
  • KITTI Cars Moderate
  • KITTI Cars Hard val
  • KITTI Cars Hard
  • KITTI Cars Easy val
  • KITTI Cars Easy
  • KITTI

78 variant names, as the archive lists them.

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