Learning Independent Instance Maps for Crowd Localization

TL;DR

This paper introduces an end-to-end crowd localization framework called Independent Instance Map segmentation (IIM), which segments crowds into non-overlapping instances, improving accuracy over traditional density and detection methods, especially in dense scenes.

Contribution

The paper proposes a novel IIM framework with a differentiable Binarization Module for improved crowd localization, handling large-scale variations and dense scenes effectively.

Findings

Outperforms state-of-the-art methods on five crowd datasets.

Improves F1-measure by 10.4% on NWPU-Crowd Localization.

Effective segmentation of independent crowd instances.

Abstract

Accurately locating each head's position in the crowd scenes is a crucial task in the field of crowd analysis. However, traditional density-based methods only predict coarse prediction, and segmentation/detection-based methods cannot handle extremely dense scenes and large-range scale-variations crowds. To this end, we propose an end-to-end and straightforward framework for crowd localization, named Independent Instance Map segmentation (IIM). Different from density maps and boxes regression, each instance in IIM is non-overlapped. By segmenting crowds into independent connected components, the positions and the crowd counts (the centers and the number of components, respectively) are obtained. Furthermore, to improve the segmentation quality for different density regions, we present a differentiable Binarization Module (BM) to output structured instance maps. BM brings two advantages into localization models: 1) adaptively learn a threshold map for different images to detect each instance more accurately; 2) directly train the model using loss on binary predictions and labels. Extensive experiments verify the proposed method is effective and outperforms the-state-of-the-art methods on the five popular crowd datasets. Significantly, IIM improves F1-measure by 10.4% on the NWPU-Crowd Localization task. The source code and pre-trained models will be released at https://github.com/taohan10200/IIM.