DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better

TL;DR

DeblurGAN-v2 introduces a fast, flexible, and high-quality GAN-based method for single image deblurring, leveraging feature pyramid networks and various backbones to achieve state-of-the-art results and real-time performance.

Contribution

It is the first to incorporate feature pyramid networks into deblurring and demonstrates flexible backbone integration for improved efficiency and quality.

Findings

Achieves 10-100x faster deblurring with lightweight backbones.

Maintains near state-of-the-art quality on benchmarks.

Effective for general image restoration tasks.

Abstract

We present a new end-to-end generative adversarial network (GAN) for single image motion deblurring, named DeblurGAN-v2, which considerably boosts state-of-the-art deblurring efficiency, quality, and flexibility. DeblurGAN-v2 is based on a relativistic conditional GAN with a double-scale discriminator. For the first time, we introduce the Feature Pyramid Network into deblurring, as a core building block in the generator of DeblurGAN-v2. It can flexibly work with a wide range of backbones, to navigate the balance between performance and efficiency. The plug-in of sophisticated backbones (e.g., Inception-ResNet-v2) can lead to solid state-of-the-art deblurring. Meanwhile, with light-weight backbones (e.g., MobileNet and its variants), DeblurGAN-v2 reaches 10-100 times faster than the nearest competitors, while maintaining close to state-of-the-art results, implying the option of real-time video deblurring. We demonstrate that DeblurGAN-v2 obtains very competitive performance on several popular benchmarks, in terms of deblurring quality (both objective and subjective), as well as efficiency. Besides, we show the architecture to be effective for general image restoration tasks too. Our codes, models and data are available at: https://github.com/KupynOrest/DeblurGANv2