Semi-Supervised Crowd Counting via Self-Training on Surrogate Tasks

TL;DR

This paper introduces a semi-supervised crowd counting approach that leverages unlabeled data to train a robust feature extractor using surrogate binary segmentation tasks and self-training, reducing annotation costs and improving performance.

Contribution

It proposes a novel semi-supervised method that trains a feature extractor with surrogate tasks and self-training, enhancing crowd counting with fewer labeled data.

Findings

Outperforms existing semi-supervised crowd counting methods

Effective use of surrogate binary segmentation tasks

Robust feature learning from unlabeled data

Abstract

Most existing crowd counting systems rely on the availability of the object location annotation which can be expensive to obtain. To reduce the annotation cost, one attractive solution is to leverage a large number of unlabeled images to build a crowd counting model in semi-supervised fashion. This paper tackles the semi-supervised crowd counting problem from the perspective of feature learning. Our key idea is to leverage the unlabeled images to train a generic feature extractor rather than the entire network of a crowd counter. The rationale of this design is that learning the feature extractor can be more reliable and robust towards the inevitable noisy supervision generated from the unlabeled data. Also, on top of a good feature extractor, it is possible to build a density map regressor with much fewer density map annotations. Specifically, we proposed a novel semi-supervised crowd counting method which is built upon two innovative components: (1) a set of inter-related binary segmentation tasks are derived from the original density map regression task as the surrogate prediction target; (2) the surrogate target predictors are learned from both labeled and unlabeled data by utilizing a proposed self-training scheme which fully exploits the underlying constraints of these binary segmentation tasks. Through experiments, we show that the proposed method is superior over the existing semisupervised crowd counting method and other representative baselines.