Spatial Uncertainty-Aware Semi-Supervised Crowd Counting



Meng, Yanda ORCID: 0000-0001-7344-2174, Zhang, Hongrun, Zhao, Yitian, Yang, Xiaoyun, Qian, Xuesheng, Huang, Xiaowei ORCID: 0000-0001-6267-0366 and Zheng, Yalin ORCID: 0000-0002-7873-0922
(2021) Spatial Uncertainty-Aware Semi-Supervised Crowd Counting. In: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021-10-10 - 2021-10-17.

[thumbnail of ICCV2021_Yanda.pdf] Text
ICCV2021_Yanda.pdf - Author Accepted Manuscript

Download (2MB) | Preview

Abstract

Semi-supervised approaches for crowd counting attract attention, as the fully supervised paradigm is expensive and laborious due to its request for a large number of images of dense crowd scenarios and their annotations. This paper proposes a spatial uncertainty-aware semi-supervised approach via regularized surrogate task (binary segmentation) for crowd counting problems. Different from existing semi-supervised learning-based crowd counting methods, to exploit the unlabeled data, our proposed spatial uncertainty-aware teacher-student framework focuses on high confident regions' information while addressing the noisy supervision from the unlabeled data in an end-to-end manner. Specifically, we estimate the spatial uncertainty maps from the teacher model's surrogate task to guide the feature learning of the main task (density regression) and the surrogate task of the student model at the same time. Besides, we introduce a simple yet effective differential transformation layer to enforce the inherent spatial consistency regularization between the main task and the surrogate task in the student model, which helps the surrogate task to yield more reliable predictions and generates high-quality uncertainty maps. Thus, our model can also address the task-level perturbation problems that occur spatial inconsistency between the primary and surrogate tasks in the student model. Experimental results on four challenging crowd counting datasets demonstrate that our method achieves superior performance to the state-of-the-art semi-supervised methods.

Item Type: Conference or Workshop Item (Unspecified)
Additional Information: Accepted by ICCV2021
Uncontrolled Keywords: cs.CV, cs.CV, cs.AI
Divisions: Faculty of Health and Life Sciences
Faculty of Health and Life Sciences > Institute of Life Courses and Medical Sciences
Depositing User: Symplectic Admin
Date Deposited: 03 Aug 2021 07:30
Last Modified: 11 Jun 2024 19:08
DOI: 10.1109/ICCV48922.2021.01526
Related URLs:
URI: https://livrepository.liverpool.ac.uk/id/eprint/3132038