Learning a Reinforced Agent for Flexible Exposure Bracketing Selection
This addresses the need for flexible exposure bracketing selection in scenarios like mobile applications, where previous methods had restrictions, though it is incremental as it builds on existing multi-exposure fusion techniques.
The paper tackles the problem of automatically selecting exposure bracketing for high dynamic range imaging by proposing EBSNet, a deep neural network trained as a reinforced agent that uses a single preview image, achieving favorable results against state-of-the-art methods.
Automatically selecting exposure bracketing (images exposed differently) is important to obtain a high dynamic range image by using multi-exposure fusion. Unlike previous methods that have many restrictions such as requiring camera response function, sensor noise model, and a stream of preview images with different exposures (not accessible in some scenarios e.g. some mobile applications), we propose a novel deep neural network to automatically select exposure bracketing, named EBSNet, which is sufficiently flexible without having the above restrictions. EBSNet is formulated as a reinforced agent that is trained by maximizing rewards provided by a multi-exposure fusion network (MEFNet). By utilizing the illumination and semantic information extracted from just a single auto-exposure preview image, EBSNet can select an optimal exposure bracketing for multi-exposure fusion. EBSNet and MEFNet can be jointly trained to produce favorable results against recent state-of-the-art approaches. To facilitate future research, we provide a new benchmark dataset for multi-exposure selection and fusion.