Latent Multi-Relation Reasoning for GAN-Prior based Image Super-Resolution
This work addresses image super-resolution for applications requiring high-quality upscaling, representing an incremental improvement over existing GAN-prior methods.
The paper tackles the problem of attribute disentanglement and unfaithful detail generation in GAN-prior based image super-resolution (SR) under large scaling factors, proposing LAREN, a latent multi-relation reasoning technique that achieves superior SR performance and consistently outperforms state-of-the-art methods across multiple benchmarks.
Recently, single image super-resolution (SR) under large scaling factors has witnessed impressive progress by introducing pre-trained generative adversarial networks (GANs) as priors. However, most GAN-Priors based SR methods are constrained by an attribute disentanglement problem in inverted latent codes which directly leads to mismatches of visual attributes in the generator layers and further degraded reconstruction. In addition, stochastic noises fed to the generator are employed for unconditional detail generation, which tends to produce unfaithful details that compromise the fidelity of the generated SR image. We design LAREN, a LAtent multi-Relation rEasoNing technique that achieves superb large-factor SR through graph-based multi-relation reasoning in latent space. LAREN consists of two innovative designs. The first is graph-based disentanglement that constructs a superior disentangled latent space via hierarchical multi-relation reasoning. The second is graph-based code generation that produces image-specific codes progressively via recursive relation reasoning which enables prior GANs to generate desirable image details. Extensive experiments show that LAREN achieves superior large-factor image SR and outperforms the state-of-the-art consistently across multiple benchmarks.