/en/tags/diffusion-models/Recent content in Diffusion Models on Mi&Bee BlogHugo -- gohugo.ioen蓝宝石的傻话Wed, 24 Jun 2026 10:00:00 +0800/en/posts/physical-world/modern-super-resolution-esrgan-diffusion/Wed, 24 Jun 2026 10:00:00 +0800/en/posts/physical-world/modern-super-resolution-esrgan-diffusion/<h2 id="from-srgan-to-esrgan-2018">From SRGAN to ESRGAN (2018)</h2> <p>SRGAN (Super-Resolution GAN) introduced Generative Adversarial Networks (GANs) to super-resolution in 2016, achieving significant visual quality improvements over traditional PSNR-optimization methods through perceptual loss. However, SRGAN still had room for improvement. <strong>ESRGAN</strong> (Enhanced Super-Resolution GAN) by Wang et al. in 2018 optimized four key directions.</p> <p>The <strong>RRDB (Residual-in-Residual Dense Block)</strong> combines the advantages of residual and dense connections. Dense connections allow each layer to access features from all preceding layers, avoiding feature redundancy, while residual connections stabilize training of deep networks. ESRGAN stacks multiple residual dense blocks, forming a Residual-in-Residual structure—using residual connections at the macro level between blocks and dense connections within each block at the micro level.</p>