HDR single-frame synthesis method based on autoencoder

• Main Authors: WU YUSONG, SHI GUOJIE, CAO JINGCHENG
• Format: Patent
• Language: Chinese; English
• Published: 28.12.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; PHYSICS; PICTORIAL COMMUNICATION, e.g. TELEVISION

The invention provides a method and a system for generating an HDR video based on HDR single-frame synthesis of an autoencoder. The method comprises the following steps: extracting frames of an LDR video file one by one; respectively inputting each LDR video frame into a pseudo high-exposure generation model and a pseudo low-exposure generation model to respectively obtain a high-exposure feature vector and a low-exposure feature vector output by a corresponding encoder, wherein the pseudo high-exposure generation model and the pseudo low-exposure generation model are used for respectively generating a high-exposure image and a low-exposure image corresponding to an input image; inputting the obtained high-exposure feature vectors and low-exposure feature vectors into a multi-exposure image fusion model to obtain processed HDR video frames, wherein the multi-exposure image fusion model performs feature extraction by using a convolutional neural network, fuses the corresponding feature vectors and reconstructs