SeedVR: Seeding Infinity in Diffusion Transformer Towards Generic Video Restoration

TL;DR

SeedVR introduces a diffusion transformer with shifted window attention for efficient, high-quality, and flexible video restoration across various resolutions and lengths, outperforming existing methods.

Contribution

The paper proposes SeedVR, a novel diffusion transformer with shifted window attention, enabling effective restoration of long, variable-resolution videos with improved efficiency and performance.

Findings

Outperforms existing video restoration methods on benchmarks

Handles arbitrary video length and resolution effectively

Achieves high fidelity and temporal consistency in restored videos

Abstract

Video restoration poses non-trivial challenges in maintaining fidelity while recovering temporally consistent details from unknown degradations in the wild. Despite recent advances in diffusion-based restoration, these methods often face limitations in generation capability and sampling efficiency. In this work, we present SeedVR, a diffusion transformer designed to handle real-world video restoration with arbitrary length and resolution. The core design of SeedVR lies in the shifted window attention that facilitates effective restoration on long video sequences. SeedVR further supports variable-sized windows near the boundary of both spatial and temporal dimensions, overcoming the resolution constraints of traditional window attention. Equipped with contemporary practices, including causal video autoencoder, mixed image and video training, and progressive training, SeedVR achieves highly-competitive performance on both synthetic and real-world benchmarks, as well as AI-generated videos. Extensive experiments demonstrate SeedVR's superiority over existing methods for generic video restoration.