SINGLE-IMAGE HDR RECONSTRUCTION BASED ON TWO-STAGE GAN STRUCTURE

Because human eyes can capture a wide range of luminance, the high dynamic range (HDR) reconstruction is to expand the low dynamic range (LDR) image to that we actually see. It is challenging to recover an HDR image from a single LDR image due to the missing information in under-/over-exposed regions. In this paper, we proposed a novel two-stage model to settle this problem. The first stage model takes the generative adversarial network (GAN) and the attention mechanism to generate the missing information in under-/over-exposed areas. The second stage is a multi-branch convolutional neural network (CNN) to fuse the multiple different exposure LDR images from the first stage to generate an HDR image. The quantitative comparisons demonstrate that our method achieves higher scores than other state-of-the-art methods. Moreover, our method gene-rates smooth and noise-free HDR images in the qualitative comparisons.