CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes

CVPR 2018  ·  Yuhong Li, Xiaofan Zhang, Deming Chen ·

We propose a network for Congested Scene Recognition called CSRNet to provide a data-driven and deep learning method that can understand highly congested scenes and perform accurate count estimation as well as present high-quality density maps. The proposed CSRNet is composed of two major components: a convolutional neural network (CNN) as the front-end for 2D feature extraction and a dilated CNN for the back-end, which uses dilated kernels to deliver larger reception fields and to replace pooling operations. CSRNet is an easy-trained model because of its pure convolutional structure. We demonstrate CSRNet on four datasets (ShanghaiTech dataset, the UCF_CC_50 dataset, the WorldEXPO'10 dataset, and the UCSD dataset) and we deliver the state-of-the-art performance. In the ShanghaiTech Part_B dataset, CSRNet achieves 47.3% lower Mean Absolute Error (MAE) than the previous state-of-the-art method. We extend the targeted applications for counting other objects, such as the vehicle in TRANCOS dataset. Results show that CSRNet significantly improves the output quality with 15.4% lower MAE than the previous state-of-the-art approach.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Crowd Counting ShanghaiTech A CSRNet MAE 68.2 # 16
Crowd Counting ShanghaiTech B CSRNet MAE 10.6 # 14
Crowd Counting TRANCOS CSRNet MAE 3.56 # 4
Crowd Counting UCF CC 50 CSRNet MAE 266.1 # 11
Crowd Counting WorldExpo’10 CSRNet Average MAE 8.6 # 7

Results from Other Papers

Task Dataset Model Metric Name Metric Value Rank Source Paper Compare
Crowd Counting Venice CSRNet MAE 35.8 # 3


No methods listed for this paper. Add relevant methods here