• Cite Icon49
  • https://doi.org/10.1007/s00138-018-0955-6Copy DOI Icon

Mixture of counting CNNs

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

This paper proposes a crowd counting method. Crowd counting is difficult because of significant appearance changes of a target which caused by density and scale changes. Conventional crowd counting methods commonly utilize one predictor (e.g., regression and multi-class classifier). However, such only one predictor can not count targets with significant appearance changes well. In this paper, we propose to predict the number of targets using multiple convolutional neural networks (CNNs) specialized to a specific appearance, and those CNNs are adaptively selected according to the appearance of a test image. By integrating the selected CNNs, the proposed method has the robustness to large appearance changes. In experiments, we confirm that the proposed method can count crowd with lower counting error than VGGNet, integration of CNNs with fixed weights and conventional counting methods. Moreover, we confirm that each CNN automatically specialized to a specific appearance (e.g., dense region and sparse region) of crowd through training of CNNs.

Similar Papers
  • Conference Article
  • Citations20

Deep Self-Paced Learning for Semi-Supervised Person Re-Identification Using Multi-View Self-Paced Clustering

  • Sep 01, 2019
  • Xiaomeng Xin +3
  • Conference Article

A crowd counting method based on multi-column dilated convolutional neural network

  • May 13, 2019
  • Weiqun Wu +4
  • Research Article
  • Citations114

A survey of crowd counting and density estimation based on convolutional neural network

  • Nov 08, 2021
  • Neurocomputing
  • Zizhu Fan +5
  • Research Article
  • Citations315

DeepTrack: Learning Discriminative Feature Representations Online for Robust Visual Tracking.

  • Dec 22, 2015
  • IEEE Transactions on Image Processing
  • Hanxi Li +2
  • Book Chapter
  • Citations98

Robust Online Visual Tracking with a Single Convolutional Neural Network

  • Jan 01, 2015
  • Hanxi Li +2
  • Research Article
  • Citations67

ST-CNN: Spatial-Temporal Convolutional Neural Network for crowd counting in videos

  • Apr 16, 2019
  • Pattern Recognition Letters
  • Yunqi Miao +3
  • Research Article
  • Citations12

Improvement of spatial resolution in a Timepix based CdTe photon counting detector using ToT method

  • Feb 24, 2018
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
  • Kyeongjin Park +6
  • Research Article
  • Citations4

Automated Documentation of Multiple Colonoscopy Quality Measures in Real-Time with Convolutional Neural Networks

  • Oct 01, 2018
  • American Journal of Gastroenterology
  • William Karnes +5
  • Conference Article
  • Citations4

Crowd counting and density estimation method based on multi-column CNN and adaptive projections onto convex sets

  • Sep 01, 2021
  • Gongda Qiu +3
  • Conference Article
  • Citations2

Weakly-Supervised Crowd Counting with Token Attention and Fusion: A Simple and Effective Baseline

  • Apr 14, 2024
  • Yi Wang +2
  • Conference Article
  • Citations58

Aspect-based Opinion Summarization with Convolutional Neural Networks

  • Jul 01, 2016
  • Haibing Wu +3
  • Research Article
  • Citations20

Real-Time CNN Training and Compression for Neural-Enhanced Adaptive Live Streaming.

  • Sep 01, 2024
  • IEEE transactions on pattern analysis and machine intelligence
  • Seunghwa Jeong +7
  • Conference Article
  • Citations1354

Cross-scene crowd counting via deep convolutional neural networks

  • Jun 01, 2015
  • Cong Zhang +3
  • Book Chapter
  • Citations2

Surveillance Based Crowd Counting via Convolutional Neural Networks

  • Jan 01, 2016
  • Damin Zhang +2
  • Book Chapter
  • Citations85

Crossing-Line Crowd Counting with Two-Phase Deep Neural Networks

  • Jan 01, 2016
  • Zhuoyi Zhao +3
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.