The lighting conditions of mobile devices could severely affect the viewing experience of the displayed image, such as loss of details and contrast, especially for high dynamic range (HDR) contents. In this paper, a content independent luminance and detail enhancement method is proposed for HDR images under ambient illumination. The perceptual luminance at the given lighting conditions is first quantified based on the human visual adaption. Combining the quantization curve and the display model, the mapping operator for the luminance channel could then be efficiently constructed. Both psychophysical evaluation and objective measurement indicate that the proposed method significantly improves the perceptual quality of HDR images under ambient illumination.

In this study, a novel illuminant color estimation framework is proposed for computational color constancy, which incorporates the high representational capacity of deep-learningbased models and the great interpretability of assumptionbased models. The well-designed building block, feature map reweight unit (ReWU), helps to achieve comparative accuracy on benchmark datasets with respect to prior state-of-the-art deep learning based models while requiring more compact model size and cheaper computational cost. In addition to local color estimation, a confidence estimation branch is also included such that the model is able to simultaneously produce point estimate and its uncertainty estimate, which provides useful clues for local estimates aggregation and multiple illumination estimation. The source code and the dataset have been made available1.