Awesome Crowd Counting

If you have any problems, suggestions or improvements, please submit the issue or PR.

Contents

• Tools
• Datasets
• Papers
• Leaderboard

Tools

• Density Map Generation from Key Points [Matlab Code] [Python Code]

Datasets

• UCF-QNRF Dataset [Link]
• ShanghaiTech Dataset [Link: Dropbox / BaiduNetdisk]
• WorldExpo'10 Dataset [Link]
• UCF CC 50 Dataset [Link]
• Mall Dataset [Link]
• UCSD Dataset [Link]
• SmartCity Dataset [Link: GoogleDrive / BaiduNetdisk]
• AHU-Crowd Dataset [Link]

Papers

arXiv papers

This section only includes the last ten papers since 2018 in arXiv.org. Previous papers will be hidden using <!--...-->. If you want to view them, please open the raw file to read the source code. Note that all unpublished arXiv papers are not included into the leaderboard of performance.

• Stacked Pooling: Improving Crowd Counting by Boosting Scale Invariance [paper][code]
• In Defense of Single-column Networks for Crowd Counting [paper]
• Perspective-Aware CNN For Crowd Counting [paper]
• Attention to Head Locations for Crowd Counting [paper]
• Crowd Counting with Density Adaption Networks [paper]
• Geometric and Physical Constraints for Head Plane Crowd Density Estimation in Videos [paper]
• Improving Object Counting with Heatmap Regulation [paper][code]
• Depth Information Guided Crowd Counting for Complex Crowd Scenes [paper]
• Structured Inhomogeneous Density Map Learning for Crowd Counting [paper]

2018

• Crowd Counting by Adaptively Fusing Predictions from an Image Pyramid (BMVC2018) [paper]
• Crowd Counting using Deep Recurrent Spatial-Aware Network (IJCAI2018) [paper]
• Top-Down Feedback for Crowd Counting Convolutional Neural Network (AAAI2018) [paper]
• [SANet] Scale Aggregation Network for Accurate and Efficient Crowd Counting (ECCV2018) [paper]
• [ic-CNN] Iterative Crowd Counting (ECCV2018) [paper]
• [CL] Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowds (ECCV2018) [paper]
• Crowd Counting with Deep Negative Correlation Learning (CVPR2018) [paper] [code]
• [IG-CNN] Divide and Grow: Capturing Huge Diversity in Crowd Images with Incrementally Growing CNN (CVPR2018) [paper]
• [BSAD] Body Structure Aware Deep Crowd Counting (TIP2018) [paper]
• [CSR] CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes (CVPR2018) [paper] [code]
• [L2R] Leveraging Unlabeled Data for Crowd Counting by Learning to Rank (CVPR2018) [paper] [code]
• [ACSCP] Crowd Counting via Adversarial Cross-Scale Consistency Pursuit (CVPR2018) [paper]
• [DecideNet] DecideNet: Counting Varying Density Crowds Through Attention Guided Detection and Density (CVPR2018) [paper]
• [DR-ResNet] A Deeply-Recursive Convolutional Network for Crowd Counting (ICASSP2018) [paper]
• [SaCNN] Crowd counting via scale-adaptive convolutional neural network (WACV2018) [paper] [code]

2017

• Generating High-Quality Crowd Density Maps using Contextual Pyramid CNNs (ICCV2017) [paper]
• Spatiotemporal Modeling for Crowd Counting in Videos (ICCV2017) [paper]
• CNN-based Cascaded Multi-task Learning of High-level Prior and Density Estimation for Crowd Counting (AVSS2017) [paper] [code]
• Switching Convolutional Neural Network for Crowd Counting (CVPR2017) [paper] [code]
• A Survey of Recent Advances in CNN-based Single Image Crowd Counting and Density Estimation (PR Letters) [paper]
• Multi-scale Convolution Neural Networks for Crowd Counting (ICIP2017) [paper] [code]

2016

• Towards perspective-free object counting with deep learning (ECCV2016) [paper] [code]
• CrowdNet: A Deep Convolutional Network for Dense Crowd Counting (ACMMM2016) [paper] [code]
• [MCNN] Single-Image Crowd Counting via Multi-Column Convolutional Neural Network (CVPR2016) [paper] [unofficial code: TensorFlow PyTorch]

2015

• COUNT Forest: CO-voting Uncertain Number of Targets using Random Forest for Crowd Density Estimation (ICCV2015) [paper]
• Cross-scene Crowd Counting via Deep Convolutional Neural Networks (CVPR2015) [paper] [code]

2013

• Multi-Source Multi-Scale Counting in Extremely Dense Crowd Images (CVPR2013) [paper]
• Crossing the Line: Crowd Counting by Integer Programming with Local Features (CVPR2013) [paper]

2012

• Feature mining for localised crowd counting (BMVC2012) [paper]

2008

• Privacy preserving crowd monitoring: Counting people without people models or tracking (CVPR 2008) [paper]

Leaderboard

The section is being continually updated. Note that some values have superscript, which indicates their source.

ShanghaiTech Part A

Year-Conference/Journal	Method	MAE	MSE	PSNR	SSIM	Model Size	Params	Pre-trained Model
2018--ECCV	SANet	67.0	104.5	-	-	-	0.91M	None
2018--ECCV	ic-CNN	69.8	117.3	-	-	-	-	None
2018--CVPR	CSR	68.2	115.0	23.79	0.76	-	16.26MSANet	VGG-16
2018--CVPR	L2R	73.6	112.0	-	-	-	-	VGG-16
2018--CVPR	ACSCP	75.7	102.7	-	-	-	5.1M	None
2016--CVPR	MCNN	110.2	173.2	21.4CSR	0.52CSR	-	0.13MSANet	None

ShanghaiTech Part B

Year-Conference/Journal	Method	MAE	MSE
2018--ECCV	SANet	8.4	13.6
2018--ECCV	ic-CNN	10.7	16.0
2018--TIP	BSAD	20.2	35.6
2018--CVPR	CSR	10.6	16.0
2018--CVPR	L2R	13.7	21.4
2018--CVPR	DecideNet	21.53	31.98
2018--CVPR	ACSCP	17.2	27.4
2016--CVPR	MCNN	26.4	41.3

UCF-QNRF

Year-Conference/Journal	Method	MAE	MSE
2018--ECCV	CL	132	191
2016--CVPR	MCNN	277CL	426CL

UCF_CC_50

Year-Conference/Journal	Method	MAE	MSE
2018--ECCV	SANet	258.4	334.9
2018--ECCV	ic-CNN	260.9	365.5
2018--TIP	BSAD	409.5	563.7
2018--CVPR	CSR	266.1	397.5
2018--CVPR	L2R	279.6	388.9
2018--CVPR	ACSCP	291.0	404.6

WorldExpo'10

Year-Conference/Journal	Method	S1	S2	S3	S4	S5	Avg.
2018--ECCV	SANet	2.6	13.2	9.0	13.3	3.0	8.2
2018--ECCV	ic-CNN	17.0	12.3	9.2	8.1	4.7	10.3
2018--TIP	BSAD	4.1	21.7	11.9	11.0	3.5	10.5
2018--CVPR	CSR	2.9	11.5	8.6	16.6	3.4	8.6
2018--CVPR	DecideNet	2.0	13.14	8.90	17.40	4.75	9.23
2018--CVPR	ACSCP	2.8	14.05	9.6	8.1	2.9	7.5

UCSD

Year-Conference/Journal	Method	MAE	MSE
2018--ECCV	SANet	1.02	1.29
2018--TIP	BSAD	1.00	1.40
2018--CVPR	CSR	1.16	1.47
2018--CVPR	ACSCP	1.04	1.35