Recently developed generative methods, including invertible rescaling network
(IRN) based and generative adversarial network (GAN) based methods, have
demonstrated exceptional performance in image rescaling. However, IRN-based
methods tend to produce over-smoothed results, while GAN-based methods easily
generate fake details, which thus hinders their real applications. To address
this issue, we propose Boundary-aware Decoupled Flow Networks (BDFlow) to
generate realistic and visually pleasing results. Unlike previous methods that
model high-frequency information as standard Gaussian distribution directly,
our BDFlow first decouples the high-frequency information into \textit{semantic
high-frequency} that adheres to a Boundary distribution and
\textit{non-semantic high-frequency} counterpart that adheres to a Gaussian
distribution. Specifically, to capture semantic high-frequency parts
accurately, we use Boundary-aware Mask (BAM) to constrain the model to produce
rich textures, while non-semantic high-frequency part is randomly sampled from
a Gaussian distribution.Comprehensive experiments demonstrate that our BDFlow
significantly outperforms other state-of-the-art methods while maintaining
lower complexity. Notably, our BDFlow improves the PSNR by $4.4$ dB and the
SSIM by $0.1$ on average over GRAIN, utilizing only 74\% of the parameters and
20\% of the computation. The code will be available at
this https URL

提出了 Boundary-aware Decoupled Flow Networks (BDFlow) 来生成逼真且视觉上令人愉悦的结果，通过将高频信息分解为 “语义高频” 和 “非语义高频” 两部分，并使用 Boundary-aware Mask (BAM) 约束模型生成丰富的纹理，实验证明 BDFlow 在保持较低复杂性的同时，显著优于其他最先进的方法。

边界感知解耦流网络用于逼真的极端缩放

Boundary-aware Decoupled Flow Networks for Realistic Extreme Rescaling

This paper presents a novel Diffusion-Wavelet (DiWa) approach for
Single-Image Super-Resolution (SISR). It leverages the strengths of Denoising
Diffusion Probabilistic Models (DDPMs) and Discrete Wavelet Transformation
(DWT). By enabling DDPMs to operate in the DWT domain, our DDPM models
effectively hallucinate high-frequency information for super-resolved images on
the wavelet spectrum, resulting in high-quality and detailed reconstructions in
image space. Quantitatively, we outperform state-of-the-art diffusion-based
SISR methods, namely SR3 and SRDiff, regarding PSNR, SSIM, and LPIPS on both
face (8x scaling) and general (4x scaling) SR benchmarks. Meanwhile, using DWT
enabled us to use fewer parameters than the compared models: 92M parameters
instead of 550M compared to SR3 and 9.3M instead of 12M compared to SRDiff.
Additionally, our method outperforms other state-of-the-art generative methods
on classical general SR datasets while saving inference time. Finally, our work
highlights its potential for various applications.

本文提出了一种新颖的扩散小波（DiWa）方法，用于单图超分辨率（SISR），能够将去噪扩散概率模型（DDPMs）的优点与离散小波变换（DWT）相结合，实现在小波频谱上为超分辨图像产生高频信息，从而在图像空间中生成高质量、详细的重建。在定量指标上，我们在面部（8 倍放大）和一般（4 倍放大）SR 基准测试中均优于当前最先进的扩散 SISR 方法，即 SR3 和 SRDiff，包括 PSNR、SSIM 和 LPIPS 等方面。同时，使用 DWT 使得我们可以使用比比较的模型更少的参数：92M 参数而不是 550M 参数，与 SR3 相比，9.3M 参数而不是 12M 参数，与 SRDiff 相比。此外，我们的方法在经典的一般 SR 数据集上优于其他最先进的生成方法，同时节省推理时间。最后，我们的工作强调了它在各种应用中的潜力。