Conditional Rgb-T Fusion For Effective Crowd Counting

Many remarkable works have been proposed to deal with distortions problems in image fusion to date. However, the spectral distortion and the spatial distortion cannot always be well addressed at the same time. To deal with this, we propose an Adaptive Feature Pyramid Network (AFPN) to efficiently embed an Adaptive Detail injection (ADI) module at different scales. Feature-domain injection gains are proposed in the ADI module to adaptively modulate spatial details information and guide a refined detail injection. Furthermore, we propose a texture loss function to further guide our model to learn details perception in each band. Experiments on QuickBird and Gaofen-1 datasets show that our method achieves superior performance and produces visually pleasing fusion images.