Improving the Learning of Multi-column Convolutional Neural Network for Crowd Counting

Abstract

Tremendous variation in the scale of people/head size is a critical problem for crowd counting. To improve the scale invariance of feature representation, recent works extensively employ Convolutional Neural Networks with multi-column structures to handle different scales and resolutions. However, due to the substantial redundant parameters in columns, existing multi-column networks invariably exhibit almost the same scale features in different columns, which severely affects counting accuracy and leads to overfitting. In this paper, we attack this problem by proposing a novel Multi-column Mutual Learning (McML) strategy. It has two main innovations: 1) A statistical network is incorporated into the multi-column framework to estimate the mutual information between columns, which can approximately indicate the scale correlation between features from different columns. By minimizing the mutual information, each column is guided to learn features with different image scales. 2) We devise a mutual learning scheme that can alternately optimize each column while keeping the other columns fixed on each mini-batch training data. With such asynchronous parameter update process, each column is inclined to learn different feature representation from others, which can efficiently reduce the parameter redundancy and improve generalization ability. More remarkably, McML can be applied to all existing multi-column networks and is end-to-end trainable. Extensive experiments on four challenging benchmarks show that McML can significantly improve the original multi-column networks and outperform the other state-of-the-art approaches.

arXiv commentACM Multimedia 2019

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Learning deep representations by mutual information estimation and maximization
• Do Deep Nets Really Need to be Deep?
• MINE: Mutual Information Neural Estimation
• Crowd Counting by Adaptively Fusing Predictions from an Image Pyramid
• Top-Down Feedback for Crowd Counting Convolutional Neural Network
• ADCrowdNet: An Attention-injective Deformable Convolutional Network for Crowd Understanding
• Stacked Pooling: Improving Crowd Counting by Boosting Scale Invariance
• Attention to Head Locations for Crowd Counting
• Context-Aware Crowd Counting
• Adaptive Scenario Discovery for Crowd Counting