Research on High Dynamic Range Image Reconstruction under Significant Motion Based on Deep Learning
The high-definition live broadcasting service is crucial in the broadcasting industry.However,challenges arise due to varying intensity of lighting and the movement of subjects,leading to issues such as unclear images and unnatural brightness in live broadcasts.This paper proposes a method based on deep learning technology for reconstructing high dynamic range images under significant motion.Experimental results demonstrate that this method not only enhances image clarity but also addresses issues like uneven brightness and motion blur,thereby further improving the user's live broadcasting viewing experience.Additionally,the proposed method significantly reduces the computational cost of the model,providing robust support for the advancement of intelligent high-definition live broadcasting technology.
High dynamic range(HDR)image reconstructionDeep learningSignificant motion imagesHSV color space
NETL
NSTL
万方数据
