Punnappurath Abhijith, Brown Michael S
IEEE Trans Pattern Anal Mach Intell. 2022 Dec;44(12):9718-9724. doi: 10.1109/TPAMI.2021.3125692. Epub 2022 Nov 7.
Imaging sensors digitize incoming scene light at a dynamic range of 10-12 bits (i.e., 1024-4096 tonal values). The sensor image is then processed onboard the camera and finally quantized to only 8 bits (i.e., 256 tonal values) to conform to prevailing encoding standards. There are a number of important applications, such as high-bit-depth displays and photo editing, where it is beneficial to recover the lost bit depth. Deep neural networks are effective at this bit-depth reconstruction task. Given the quantized low-bit-depth image as input, existing deep learning methods employ a single-shot approach that attempts to either (1) directly estimate the high-bit-depth image, or (2) directly estimate the residual between the high- and low-bit-depth images. In contrast, we propose a training and inference strategy that recovers the residual image bitplane-by-bitplane. Our bitplane-wise learning framework has the advantage of allowing for multiple levels of supervision during training and is able to obtain state-of-the-art results using a simple network architecture. We test our proposed method extensively on several image datasets and demonstrate an improvement from 0.5dB to 2.3dB PSNR over prior methods depending on the quantization level.
成像传感器以10 - 12位的动态范围（即1024 - 4096个色调值）对入射场景光进行数字化处理。然后，传感器图像在相机板载上进行处理，最后量化为仅8位（即256个色调值），以符合现行的编码标准。在许多重要应用中，如高比特深度显示和照片编辑，恢复丢失的比特深度是有益的。深度神经网络在这项比特深度重建任务中很有效。以量化后的低比特深度图像作为输入，现有的深度学习方法采用单次方法，试图要么（1）直接估计高比特深度图像，要么（2）直接估计高比特深度图像和低比特深度图像之间的残差。相比之下，我们提出了一种逐比特平面恢复残差图像的训练和推理策略。我们的逐比特平面学习框架的优点是在训练期间允许进行多级监督，并且能够使用简单的网络架构获得最优结果。我们在几个图像数据集上广泛测试了我们提出的方法，并根据量化级别表明，与先前方法相比，峰值信噪比（PSNR）提高了0.5dB至2.3dB。
