Video deblurring method and apparatus, storage medium, and electronic apparatus

A video deblurring apparatus, comprising:

• processing circuitry configured to
• acquire N continuous image frames from a video clip, the N being a positive integer, and the N continuous image frames including a blurry image frame to be processed;
• perform three-dimensional (3D) convolution processing on the N continuous image frames with a generative adversarial network model that includes a generative network model and a discriminative network model, to acquire spatio-temporal information corresponding to the blurry image frame, the spatio-temporal information including spatial feature information of the blurry image frame, and temporal feature information between the blurry image frame and a neighboring image frame of the N continuous image frames, the generative network model being trained based on a first reconstruction loss function, a first adversarial loss function, and a fused loss function that is determined based on a second reconstruction loss function and a second adversarial loss function, the second reconstruction loss function being reacquired through the generative network model and the second adversarial loss function being reacquired through the discriminative network model;
• and
• perform deblurring processing on the blurry image frame by using the spatio-temporal information corresponding to the blurry image frame through the generative adversarial network model, to output a sharp image frame.

A non-transitory computer-readable storage medium storing instructions which when executed by at least one processor cause the at least one processor to perform:

• acquiring N continuous image frames from a video clip, the N being a positive integer, and the N continuous image frames including a blurry image frame to be processed;
• performing three-dimensional (3D) convolution processing on the N continuous image frames with a generative adversarial network model that includes a generative network model and a discriminative network model, to acquire spatio-temporal information corresponding to the blurry image frame, the spatio-temporal information including spatial feature information of the blurry image frame, and temporal feature information between the blurry image frame and a neighboring image frame of the N continuous image frames, the generative network model being trained based on a first reconstruction loss function, a first adversarial loss function, and a fused loss function that is determined based on a second reconstruction loss function and a second adversarial loss function, the second reconstruction loss function being reacquired through the generative network model and the second adversarial loss function being reacquired through the discriminative network model;
• and
• performing deblurring processing on the blurry image frame by using the spatio-temporal information corresponding to the blurry image frame through the generative adversarial network model, to output a sharp image frame.