Crowd counting on static images is a challenging problem due to scalevariations. Recently deep neural networks have been shown to be effective inthis task. However, existing neural-networks-based methods often use themulti-column or multi-network model to extract the scale-relevant features,which is more complicated for optimization and computation wasting. To thisend, we propose a novel multi-scale convolutional neural network (MSCNN) forsingle image crowd counting. Based on the multi-scale blobs, the network isable to generate scale-relevant features for higher crowd counting performancesin a single-column architecture, which is both accuracy and cost effective forpractical applications. Complemental results show that our method outperformsthe state-of-the-art methods on both accuracy and robustness with far lessnumber of parameters.
展开▼