Densely Connected Convolutional Networks for Multi-Exposure Fusion
Yue Que, Hyo Jong Lee · 2018
In this paper, we propose a novel deep learning network for multi-exposure fusion problem. In contrast to conventional convolutional networks, our feature extraction layers are densely connected convolutional networks (DenseNet), in which the output of each layer is connected to every other layer. We attempt to use this architecture to get more useful features from source images in feature extraction process. These low-level features of input image pairs are fused for reconstructing the final result. The proposed approach uses a novel DenseNet architecture trained to learn the fusion operation without reference ground truth image. Compared with existing fusion methods, the proposed fusion method achieves better performance.