Virtual Classification: Modulating Domain-Specific Knowledge for Multidomain Crowd Counting
This addresses the problem of domain bias for researchers and practitioners in computer vision working on crowd counting across multiple datasets, representing an incremental improvement with a novel method.
The paper tackles domain bias in multidomain crowd counting by proposing MDKNet, which uses modulating techniques like Instance-specific Batch Normalization and a Domain-guided Virtual Classifier to balance distributions across datasets, achieving state-of-the-art results on benchmarks such as Shanghai-tech A/B, QNRF, and NWPU.
Multidomain crowd counting aims to learn a general model for multiple diverse datasets. However, deep networks prefer modeling distributions of the dominant domains instead of all domains, which is known as domain bias. In this study, we propose a simple-yet-effective Modulating Domain-specific Knowledge Network (MDKNet) to handle the domain bias issue in multidomain crowd counting. MDKNet is achieved by employing the idea of `modulating', enabling deep network balancing and modeling different distributions of diverse datasets with little bias. Specifically, we propose an Instance-specific Batch Normalization (IsBN) module, which serves as a base modulator to refine the information flow to be adaptive to domain distributions. To precisely modulating the domain-specific information, the Domain-guided Virtual Classifier (DVC) is then introduced to learn a domain-separable latent space. This space is employed as an input guidance for the IsBN modulator, such that the mixture distributions of multiple datasets can be well treated. Extensive experiments performed on popular benchmarks, including Shanghai-tech A/B, QNRF and NWPU, validate the superiority of MDKNet in tackling multidomain crowd counting and the effectiveness for multidomain learning. Code is available at \url{https://github.com/csguomy/MDKNet}.