Abstract:Content Aiming at the problems of single convolution model, insufficient Receptive field and inaccurate feedback information of single discriminant network in current face image super-resolution reconstruction algorithm, an algorithm based on adaptive convolution and joint Loss function is designed. The model uses a generation adversarial network architecture. On the generator side, adaptive convolution is used to construct dual path residual blocks and further form efficient residual groups. It can independently learn feature weights extracted under different receptive fields and supplement missing information from a single branch. It uses subpixel convolution layers to complete quadruple reconstruction of face images. In terms of discriminators, Vgg and U-net architecture networks are used as dual discriminant networks, and dual discriminant results are used to calculate adversarial losses. The losses, content losses, and perceptual losses form a joint loss function. Experiments on the Celeba dataset show that compared with RWSA, this algorithm improves PSNR by 1.166 dB, SSIM by 0.037, LPIPS by 0.033, and PI by 0.119, compared with other mainstream algorithms, it has advantages in image detail clarity.