Crowd Counting with Deep Structured Scale Integration Network

TL;DR

This paper introduces DSSINet, a novel deep network that effectively handles scale variation in crowd counting by using structured feature learning and a specialized loss function, leading to significant accuracy improvements.

Contribution

The paper proposes a structured feature enhancement module with CRFs and a dilated multiscale structural similarity loss for improved crowd density estimation.

Findings

Achieves 9.5% error reduction on Shanghaitech dataset.

Achieves 24.9% error reduction on UCF-QNRF dataset.

Demonstrates superior performance over state-of-the-art methods.

Abstract

Automatic estimation of the number of people in unconstrained crowded scenes is a challenging task and one major difficulty stems from the huge scale variation of people. In this paper, we propose a novel Deep Structured Scale Integration Network (DSSINet) for crowd counting, which addresses the scale variation of people by using structured feature representation learning and hierarchically structured loss function optimization. Unlike conventional methods which directly fuse multiple features with weighted average or concatenation, we first introduce a Structured Feature Enhancement Module based on conditional random fields (CRFs) to refine multiscale features mutually with a message passing mechanism. In this module, each scale-specific feature is considered as a continuous random variable and passes complementary information to refine the features at other scales. Second, we utilize a Dilated Multiscale Structural Similarity loss to enforce our DSSINet to learn the local correlation of people's scales within regions of various size, thus yielding high-quality density maps. Extensive experiments on four challenging benchmarks well demonstrate the effectiveness of our method. Specifically, our DSSINet achieves improvements of 9.5% error reduction on Shanghaitech dataset and 24.9% on UCF-QNRF dataset against the state-of-the-art methods.