Variational Attention: Propagating Domain-Specific Knowledge for Multi-Domain Learning in Crowd Counting
This addresses the challenge of laborious labeling in crowd counting by enabling multi-domain learning, though it is incremental as it builds on existing attention mechanisms.
The paper tackles the problem of learning a general crowd counting model from multiple datasets with diverse densities and scenes, proposing a Domain-specific Knowledge Propagating Network (DKPNet) that achieves state-of-the-art performance on datasets like ShanghaiTech A/B, UCF-QNRF, and NWPU.
In crowd counting, due to the problem of laborious labelling, it is perceived intractability of collecting a new large-scale dataset which has plentiful images with large diversity in density, scene, etc. Thus, for learning a general model, training with data from multiple different datasets might be a remedy and be of great value. In this paper, we resort to the multi-domain joint learning and propose a simple but effective Domain-specific Knowledge Propagating Network (DKPNet)1 for unbiasedly learning the knowledge from multiple diverse data domains at the same time. It is mainly achieved by proposing the novel Variational Attention(VA) technique for explicitly modeling the attention distributions for different domains. And as an extension to VA, Intrinsic Variational Attention(InVA) is proposed to handle the problems of over-lapped domains and sub-domains. Extensive experiments have been conducted to validate the superiority of our DKPNet over several popular datasets, including ShanghaiTech A/B, UCF-QNRF and NWPU.