Improving Dense Crowd Counting Convolutional Neural Networks using Inverse k-Nearest Neighbor Maps and Multiscale Upsampling
This work addresses crowd counting for safety and event management, offering incremental improvements through novel labeling and upsampling techniques applicable to existing architectures.
The paper tackles the problem of automated dense crowd counting by proposing an alternative labeling scheme using inverse k-nearest neighbor (i$k$NN) maps and a new network architecture with multiscale upsampling, which improves accuracy compared to existing state-of-the-art methods.
Gatherings of thousands to millions of people frequently occur for an enormous variety of events, and automated counting of these high-density crowds is useful for safety, management, and measuring significance of an event. In this work, we show that the regularly accepted labeling scheme of crowd density maps for training deep neural networks is less effective than our alternative inverse k-nearest neighbor (i$k$NN) maps, even when used directly in existing state-of-the-art network structures. We also provide a new network architecture MUD-i$k$NN, which uses multi-scale upsampling via transposed convolutions to take full advantage of the provided i$k$NN labeling. This upsampling combined with the i$k$NN maps further improves crowd counting accuracy. Our new network architecture performs favorably in comparison with the state-of-the-art. However, our labeling and upsampling techniques are generally applicable to existing crowd counting architectures.