LDC-Net: A Unified Framework for Localization, Detection and Counting in Dense Crowds
arXiv:2110.04727
Abstract
The rapid development in visual crowd analysis shows a trend to count people by positioning or even detecting, rather than simply summing a density map. It also enlightens us back to the essence of the field, detection to count, which can give more abundant crowd information and has more practical applications. However, some recent work on crowd localization and detection has two limitations: 1) The typical detection methods can not handle the dense crowds and a large variation in scale; 2) The density map heuristic methods suffer from performance deficiency in position and box prediction, especially in high density or large-size crowds. In this paper, we devise a tailored baseline for dense crowds location, detection, and counting from a new perspective, named as LDC-Net for convenience, which has the following features: 1) A strong but minimalist paradigm to detect objects by only predicting a location map and a size map, which endows an ability to detect in a scene with any capacity ( persons); 2) Excellent cross-scale ability in facing a large variation, such as the head ranging in pixels; 3) Achieve superior performance in location and box prediction tasks, as well as a competitive counting performance compared with the density-based methods. Finally, the source code and pre-trained models will be released.
arXiv comment17 Pages, 11 figures
References in corpus (21)
- Adam: A Method for Stochastic Optimization
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- NWPU-Crowd: A Large-Scale Benchmark for Crowd Counting and Localization
- RetinaFace: Single-stage Dense Face Localisation in the Wild
- Distribution Matching for Crowd Counting
- CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes
- Fully Convolutional Neural Networks for Crowd Segmentation
- From Facial Parts Responses to Face Detection: A Deep Learning Approach
- Switching Convolutional Neural Network for Crowd Counting
- Bayesian Loss for Crowd Count Estimation with Point Supervision
- Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowds
- PyramidBox++: High Performance Detector for Finding Tiny Face
- C^3 Framework: An Open-source PyTorch Code for Crowd Counting
- Learning Independent Instance Maps for Crowd Localization
- Learning Spatial Awareness to Improve Crowd Counting
- Context-aware CNNs for person head detection
- Adaptive Mixture Regression Network with Local Counting Map for Crowd Counting
- Point in, Box out: Beyond Counting Persons in Crowds
- Localization in the Crowd with Topological Constraints
- Locality-constrained Spatial Transformer Network for Video Crowd Counting