Switching Convolutional Neural Network for Crowd Counting

Abstract

We propose a novel crowd counting model that maps a given crowd scene to its density. Crowd analysis is compounded by myriad of factors like inter-occlusion between people due to extreme crowding, high similarity of appearance between people and background elements, and large variability of camera view-points. Current state-of-the art approaches tackle these factors by using multi-scale CNN architectures, recurrent networks and late fusion of features from multi-column CNN with different receptive fields. We propose switching convolutional neural network that leverages variation of crowd density within an image to improve the accuracy and localization of the predicted crowd count. Patches from a grid within a crowd scene are relayed to independent CNN regressors based on crowd count prediction quality of the CNN established during training. The independent CNN regressors are designed to have different receptive fields and a switch classifier is trained to relay the crowd scene patch to the best CNN regressor. We perform extensive experiments on all major crowd counting datasets and evidence better performance compared to current state-of-the-art methods. We provide interpretable representations of the multichotomy of space of crowd scene patches inferred from the switch. It is observed that the switch relays an image patch to a particular CNN column based on density of crowd.

Version 2: Added references. Corrected Typos. Published at CVPR 2017, Honolulu, Hawaii. The first two authors contributed equally. For Project Page see http://val.serc.iisc.ernet.in/crowdcnn/

References in corpus (5)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Going Deeper with Convolutions
• CrowdNet: A Deep Convolutional Network for Dense Crowd Counting
• End-to-end people detection in crowded scenes
• SwiDeN : Convolutional Neural Networks For Depiction Invariant Object Recognition

Cited by in corpus (27)

• NWPU-Crowd: A Large-Scale Benchmark for Crowd Counting and Localization
• Context-aware Deep Feature Compression for High-speed Visual Tracking
• CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes
• Crowd Counting by Adaptively Fusing Predictions from an Image Pyramid
• Crowd counting with crowd attention convolutional neural network
• CCTrans: Simplifying and Improving Crowd Counting with Transformer
• Crowd Counting using Deep Recurrent Spatial-Aware Network
• From Open Set to Closed Set: Counting Objects by Spatial Divide-and-Conquer
• Stacked Pooling: Improving Crowd Counting by Boosting Scale Invariance
• Crowd counting via scale-adaptive convolutional neural network
• FCHD: Fast and accurate head detection in crowded scenes
• Leveraging Unlabeled Data for Crowd Counting by Learning to Rank
• Attention to Head Locations for Crowd Counting
• Structured Inhomogeneous Density Map Learning for Crowd Counting
• Crowd Counting with Density Adaption Networks
• Learning Short-Cut Connections for Object Counting
• In Defense of Single-column Networks for Crowd Counting
• Divide and Grow: Capturing Huge Diversity in Crowd Images with Incrementally Growing CNN
• Multi-Resolution Fusion and Multi-scale Input Priors Based Crowd Counting
• LDC-Net: A Unified Framework for Localization, Detection and Counting in Dense Crowds
• Adaptive Weighting Multi-Field-of-View CNN for Semantic Segmentation in Pathology
• SOFA-Net: Second-Order and First-order Attention Network for Crowd Counting
• Towards Unsupervised Crowd Counting via Regression-Detection Bi-knowledge Transfer
• DENet: A Universal Network for Counting Crowd with Varying Densities and Scales
• Crowd Transformer Network
• A-CCNN: adaptive ccnn for density estimation and crowd counting
• Recurrent Distillation based Crowd Counting