DarkVisionNet: Low-Light Imaging via RGB-NIR Fusion with Deep Inconsistency Prior
This work addresses low-light imaging for computer vision applications, offering a novel method to reduce visual artifacts, though it is incremental as it builds on existing RGB-NIR fusion techniques.
The paper tackles the problem of low-light imaging by proposing DarkVisionNet, an RGB-NIR fusion algorithm that uses a Deep Inconsistency Prior to handle structure inconsistencies amplified by noise, resulting in high-quality images with significant improvements in PSNR and SSIM, especially in extremely low-light conditions.
RGB-NIR fusion is a promising method for low-light imaging. However, high-intensity noise in low-light images amplifies the effect of structure inconsistency between RGB-NIR images, which fails existing algorithms. To handle this, we propose a new RGB-NIR fusion algorithm called Dark Vision Net (DVN) with two technical novelties: Deep Structure and Deep Inconsistency Prior (DIP). The Deep Structure extracts clear structure details in deep multiscale feature space rather than raw input space, which is more robust to noisy inputs. Based on the deep structures from both RGB and NIR domains, we introduce the DIP to leverage the structure inconsistency to guide the fusion of RGB-NIR. Benefiting from this, the proposed DVN obtains high-quality lowlight images without the visual artifacts. We also propose a new dataset called Dark Vision Dataset (DVD), consisting of aligned RGB-NIR image pairs, as the first public RGBNIR fusion benchmark. Quantitative and qualitative results on the proposed benchmark show that DVN significantly outperforms other comparison algorithms in PSNR and SSIM, especially in extremely low light conditions.