SOTAVerified

Pedestrian Detection

Pedestrian detection is the task of detecting pedestrians from a camera.

Further state-of-the-art results (e.g. on the KITTI dataset) can be found at 3D Object Detection.

( Image credit: High-level Semantic Feature Detection: A New Perspective for Pedestrian Detection )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 401–438 of 438 papers

TitleStatusHype
Basis Mapping Based Boosting for Object Detection—0
Filtered Feature Channels for Pedestrian Detection—0
Learning Scene-Specific Pedestrian Detectors Without Real Data—0
Boosting-like Deep Learning For Pedestrian Detection—0
Convolutional Channel FeaturesCode0
MOTChallenge 2015: Towards a Benchmark for Multi-Target TrackingCode0
Fast keypoint detection in video sequences—0
Exploring Human Vision Driven Features for Pedestrian Detection—0
Filtered Channel Features for Pedestrian Detection—0
Taking a Deeper Look at Pedestrians—0
Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices—0
Road Detection via On--line Label Transfer—0
Local Decorrelation For Improved Pedestrian Detection—0
Pedestrian Detection aided by Deep Learning Semantic Tasks—0
Ten Years of Pedestrian Detection, What Have We Learned?—0
Pedestrian Detection with Spatially Pooled Features and Structured Ensemble Learning—0
Hierarchical Adaptive Structural SVM for Domain Adaptation—0
Spatiotemporal Stacked Sequential Learning for Pedestrian Detection—0
Strengthening the Effectiveness of Pedestrian Detection with Spatially Pooled Features—0
Why do linear SVMs trained on HOG features perform so well?Code0
Switchable Deep Network for Pedestrian Detection—0
Pedestrian Detection in Low-resolution Imagery by Learning Multi-scale Intrinsic Motion Structures (MIMS)—0
Informed Haar-like Features Improve Pedestrian Detection—0
Confidence-Rated Multiple Instance Boosting for Object Detection—0
Word Channel Based Multiscale Pedestrian Detection Without Image Resizing and Using Only One Classifier—0
Layered Logic Classifiers: Exploring the `And' and `Or' Relations—0
K-Tangent Spaces on Riemannian Manifolds for Improved Pedestrian Detection—0
Efficient pedestrian detection by directly optimize the partial area under the ROC curve—0
Fast Multiple-Part Based Object Detection Using KD-Ferns—0
Single-Pedestrian Detection Aided by Multi-pedestrian Detection—0
Modeling Mutual Visibility Relationship in Pedestrian Detection—0
Exploring Weak Stabilization for Motion Feature Extraction—0
Optimized Pedestrian Detection for Multiple and Occluded People—0
Robust Multi-resolution Pedestrian Detection in Traffic Scenes—0
Context-Sensitive Decision Forests for Object Detection—0
Pedestrian Detection with Unsupervised Multi-Stage Feature Learning—0
Simultaneous Object Detection and Ranking with Weak Supervision—0
Joint Cascade Optimization Using A Product Of Boosted Classifiers—0
Show:102550
← PrevPage 9 of 9Next →
All datasetsCaltechCityPersonsLLVIPDVTODTJU-Ped-trafficTJU-Ped-campusCVC14Caltech Pedestrian DatasetMMPD-Dataset

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1UniHCP (FT)Heavy MR^-227.2—Unverified
2LDCFReasonable Miss Rate24.8—Unverified
3AlexNetReasonable Miss Rate23.3—Unverified
4TA-CNNReasonable Miss Rate20.9—Unverified
5Checkerboards+Reasonable Miss Rate17.1—Unverified
6NNNFReasonable Miss Rate16.2—Unverified
7Part-level CNN + saliency and bounding box alignmentReasonable Miss Rate12.4—Unverified
8CompACT-DeepReasonable Miss Rate11.75—Unverified
9MCFReasonable Miss Rate10.4—Unverified
10MS-CNNReasonable Miss Rate9.95—Unverified
#ModelMetricClaimedVerifiedStatus
1ACSP + EuroCity PersonsHeavy MR^-242.5—Unverified
2TLLReasonable MR^-215.5—Unverified
3FRCNNReasonable MR^-215.4—Unverified
4FRCNN+SegReasonable MR^-214.8—Unverified
5TLL+MRFReasonable MR^-214.4—Unverified
6RepLossReasonable MR^-213.2—Unverified
7OR-CNNReasonable MR^-212.8—Unverified
8ALFNetReasonable MR^-212—Unverified
9CSP (with offset) + ResNet-50Reasonable MR^-211—Unverified
10NOH-NMSReasonable MR^-210.8—Unverified
#ModelMetricClaimedVerifiedStatus
1INSANetlog average miss rate4.43—Unverified
2MMPedestronAP0.73—Unverified
3CAFF-DINOAP0.69—Unverified
4CFTAP0.64—Unverified
5UniRGB-IRAP0.63—Unverified
6RSDetAP0.61—Unverified
7CMXAP0.6—Unverified
8CSSAAP0.59—Unverified
9GAFFAP0.56—Unverified
10Halfway FusionAP0.55—Unverified
#ModelMetricClaimedVerifiedStatus
1YOLOv6 (Thermal) mAP84.4—Unverified
2YOLOv3 (Thermal) mAP82.7—Unverified
3CFT mAP82.7—Unverified
4CMX mAP81.6—Unverified
5YOLOv7 (Thermal)mAP77.8—Unverified
6YOLOv6 (Visible) mAP38.1—Unverified
7YOLOv7 (Visible)mAP35.3—Unverified
8YOLOv3 (Visible) mAP34.5—Unverified
#ModelMetricClaimedVerifiedStatus
1FCOSR (miss rate)24.35—Unverified
2RetinaNetR (miss rate)23.89—Unverified
3FPNR (miss rate)22.3—Unverified
4CrowdDetR (miss rate)20.82—Unverified
5EGCLR (miss rate)19.73—Unverified
6LSFMR (miss rate)18.7—Unverified
#ModelMetricClaimedVerifiedStatus
1RetinaNetR (miss rate)34.73—Unverified
2FCOSR (miss rate)31.89—Unverified
3CrowdDetR (miss rate)25.73—Unverified
4EGCLR (miss rate)24.84—Unverified
#ModelMetricClaimedVerifiedStatus
1CFTAP5078.2—Unverified
2CMXAP5068.9—Unverified
#ModelMetricClaimedVerifiedStatus
1LSFMMR0.87—Unverified
#ModelMetricClaimedVerifiedStatus
1MMPedestronbox mAP79—Unverified