ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1902.05379
10
23

Improving Dense Crowd Counting Convolutional Neural Networks using Inverse k-Nearest Neighbor Maps and Multiscale Upsampling

31 January 2019
Greg Olmschenk
Hao Tang
Zhigang Zhu
ArXivPDFHTML
Abstract

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 (ikkkNN) maps, even when used directly in existing state-of-the-art network structures. We also provide a new network architecture MUD-ikkkNN, which uses multi-scale upsampling via transposed convolutions to take full advantage of the provided ikkkNN labeling. This upsampling combined with the ikkkNN 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.

View on arXiv
Comments on this paper