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 Added Most Hyped Most Active Needs Verification Most Verified

Showing 326–350 of 438 papers

Title	Date	Tasks	Status
Multispectral Pedestrian Detection via Simultaneous Detection and Segmentation	Aug 14, 2018	Autonomous DrivingMultispectral Object Detection	CodeCode Available
Unsupervised Hard Example Mining from Videos for Improved Object Detection	Aug 13, 2018	Face Detectionobject-detection	—Unverified
SAM-RCNN: Scale-Aware Multi-Resolution Multi-Channel Pedestrian Detection	Aug 7, 2018	Pedestrian DetectionRegion Proposal	—Unverified
Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd	Jul 23, 2018	Pedestrian Detection	—Unverified
Person Search via A Mask-Guided Two-Stream CNN Model	Jul 21, 2018	Pedestrian DetectionPerson Re-Identification	—Unverified
Convolutional Neural Networks for Aerial Multi-Label Pedestrian Detection	Jul 16, 2018	Action DetectionObject	—Unverified
Is the Pedestrian going to Cross? Answering by 2D Pose Estimation	Jul 15, 2018	2D Pose EstimationPedestrian Detection	—Unverified
Object Detection with Deep Learning: A Review	Jul 15, 2018	Deep LearningFace Detection	—Unverified
Robustness Analysis of Pedestrian Detectors for Surveillance	Jul 12, 2018	Pedestrian Detection	CodeCode Available
Small-scale Pedestrian Detection Based on Somatic Topology Localization and Temporal Feature Aggregation	Jul 4, 2018	Pedestrian Detection	—Unverified
A Content-Based Late Fusion Approach Applied to Pedestrian Detection	Jun 8, 2018	Pedestrian Detection	—Unverified
Improving Occlusion and Hard Negative Handling for Single-Stage Pedestrian Detectors	Jun 1, 2018	Occlusion HandlingPedestrian Detection	—Unverified
Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning	Jun 1, 2018	One-Shot LearningPedestrian Detection	—Unverified
Occluded Pedestrian Detection Through Guided Attention in CNNs	Jun 1, 2018	Pedestrian Detection	—Unverified
WILDTRACK: A Multi-Camera HD Dataset for Dense Unscripted Pedestrian Detection	Jun 1, 2018	multi-view detectionMultiview Detection	—Unverified
Disparity Sliding Window: Object Proposals From Disparity Images	May 17, 2018	Objectobject-detection	CodeCode Available
Fused Deep Neural Networks for Efficient Pedestrian Detection	May 2, 2018	Ensemble LearningGeneral Classification	—Unverified
PCN: Part and Context Information for Pedestrian Detection with CNNs	Apr 12, 2018	Occlusion HandlingPedestrian Detection	—Unverified
Pedestrian-Synthesis-GAN: Generating Pedestrian Data in Real Scene and Beyond	Apr 5, 2018	Generative Adversarial NetworkPedestrian Detection	CodeCode Available
Exploring Multi-Branch and High-Level Semantic Networks for Improving Pedestrian Detection	Apr 3, 2018	object-detectionObject Detection	—Unverified
End-to-End Detection and Re-identification Integrated Net for Person Search	Apr 2, 2018	Pedestrian DetectionPerson Search	—Unverified
Illumination-aware Faster R-CNN for Robust Multispectral Pedestrian Detection	Mar 14, 2018	Multispectral Object DetectionPedestrian Detection	—Unverified
Fusion of Multispectral Data Through Illumination-aware Deep Neural Networks for Pedestrian Detection	Feb 27, 2018	Autonomous DrivingMultispectral Object Detection	—Unverified
Unsupervised Deep Domain Adaptation for Pedestrian Detection	Feb 9, 2018	Domain AdaptationPedestrian Detection	—Unverified
A High-Performance HOG Extractor on FPGA	Jan 12, 2018	Pedestrian DetectionVocal Bursts Intensity Prediction	—Unverified

Show:10 25 50

← PrevPage 14 of 18Next →

All datasets Caltech CityPersons LLVIP DVTOD TJU-Ped-traffic TJU-Ped-campus CVC14 Caltech Pedestrian Dataset MMPD-Dataset

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	UniHCP (FT)	Heavy MR^-2	27.2	—	Unverified
2	LDCF	Reasonable Miss Rate	24.8	—	Unverified
3	AlexNet	Reasonable Miss Rate	23.3	—	Unverified
4	TA-CNN	Reasonable Miss Rate	20.9	—	Unverified
5	Checkerboards+	Reasonable Miss Rate	17.1	—	Unverified
6	NNNF	Reasonable Miss Rate	16.2	—	Unverified
7	Part-level CNN + saliency and bounding box alignment	Reasonable Miss Rate	12.4	—	Unverified
8	CompACT-Deep	Reasonable Miss Rate	11.75	—	Unverified
9	MCF	Reasonable Miss Rate	10.4	—	Unverified
10	MS-CNN	Reasonable Miss Rate	9.95	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ACSP + EuroCity Persons	Heavy MR^-2	42.5	—	Unverified
2	TLL	Reasonable MR^-2	15.5	—	Unverified
3	FRCNN	Reasonable MR^-2	15.4	—	Unverified
4	FRCNN+Seg	Reasonable MR^-2	14.8	—	Unverified
5	TLL+MRF	Reasonable MR^-2	14.4	—	Unverified
6	RepLoss	Reasonable MR^-2	13.2	—	Unverified
7	OR-CNN	Reasonable MR^-2	12.8	—	Unverified
8	ALFNet	Reasonable MR^-2	12	—	Unverified
9	CSP (with offset) + ResNet-50	Reasonable MR^-2	11	—	Unverified
10	NOH-NMS	Reasonable MR^-2	10.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	INSANet	log average miss rate	4.43	—	Unverified
2	MMPedestron	AP	0.73	—	Unverified
3	CAFF-DINO	AP	0.69	—	Unverified
4	CFT	AP	0.64	—	Unverified
5	UniRGB-IR	AP	0.63	—	Unverified
6	RSDet	AP	0.61	—	Unverified
7	CMX	AP	0.6	—	Unverified
8	CSSA	AP	0.59	—	Unverified
9	GAFF	AP	0.56	—	Unverified
10	Halfway Fusion	AP	0.55	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	YOLOv6 (Thermal)	mAP	84.4	—	Unverified
2	CFT	mAP	82.7	—	Unverified
3	YOLOv3 (Thermal)	mAP	82.7	—	Unverified
4	CMX	mAP	81.6	—	Unverified
5	YOLOv7 (Thermal)	mAP	77.8	—	Unverified
6	YOLOv6 (Visible)	mAP	38.1	—	Unverified
7	YOLOv7 (Visible)	mAP	35.3	—	Unverified
8	YOLOv3 (Visible)	mAP	34.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	FCOS	R (miss rate)	24.35	—	Unverified
2	RetinaNet	R (miss rate)	23.89	—	Unverified
3	FPN	R (miss rate)	22.3	—	Unverified
4	CrowdDet	R (miss rate)	20.82	—	Unverified
5	EGCL	R (miss rate)	19.73	—	Unverified
6	LSFM	R (miss rate)	18.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	RetinaNet	R (miss rate)	34.73	—	Unverified
2	FCOS	R (miss rate)	31.89	—	Unverified
3	CrowdDet	R (miss rate)	25.73	—	Unverified
4	EGCL	R (miss rate)	24.84	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CFT	AP50	78.2	—	Unverified
2	CMX	AP50	68.9	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LSFM	MR	0.87	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MMPedestron	box mAP	79	—	Unverified