Rethinking Diffusion Model-Based Video Super-Resolution: Leveraging Dense Guidance from Aligned Features

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Jingyi, Zheng, Meisong, Chen, Ying, Qiao, Minglang, Deng, Xin, Xu, Mai
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917363973095424
author Xu, Jingyi
Zheng, Meisong
Chen, Ying
Qiao, Minglang
Deng, Xin
Xu, Mai
author_facet Xu, Jingyi
Zheng, Meisong
Chen, Ying
Qiao, Minglang
Deng, Xin
Xu, Mai
contents Diffusion model (DM) based Video Super-Resolution (VSR) approaches achieve impressive perceptual quality. However, they suffer from error accumulation, spatial artifacts, and a trade-off between perceptual quality and fidelity, primarily caused by inaccurate alignment and insufficient compensation between video frames. In this paper, within the DM-based VSR pipeline, we revisit the role of alignment and compensation between adjacent video frames and reveal two crucial observations: (a) the feature domain is better suited than the pixel domain for information compensation due to its stronger spatial and temporal correlations, and (b) warping at an upscaled resolution better preserves high-frequency information, but this benefit is not necessarily monotonic. Therefore, we propose a novel Densely Guided diffusion model with Aligned Features for Video Super-Resolution (DGAF-VSR), with an Optical Guided Warping Module (OGWM) to maintain high-frequency details in the aligned features and a Feature-wise Temporal Condition Module (FTCM) to deliver dense guidance in the feature domain. Extensive experiments on synthetic and real-world datasets demonstrate that DGAF-VSR surpasses state-of-the-art methods in key aspects of VSR, including perceptual quality (35.82\% DISTS reduction), fidelity (0.20 dB PSNR gain), and temporal consistency (30.37\% tLPIPS reduction).
format Preprint
id arxiv_https___arxiv_org_abs_2511_16928
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rethinking Diffusion Model-Based Video Super-Resolution: Leveraging Dense Guidance from Aligned Features
Xu, Jingyi
Zheng, Meisong
Chen, Ying
Qiao, Minglang
Deng, Xin
Xu, Mai
Computer Vision and Pattern Recognition
Diffusion model (DM) based Video Super-Resolution (VSR) approaches achieve impressive perceptual quality. However, they suffer from error accumulation, spatial artifacts, and a trade-off between perceptual quality and fidelity, primarily caused by inaccurate alignment and insufficient compensation between video frames. In this paper, within the DM-based VSR pipeline, we revisit the role of alignment and compensation between adjacent video frames and reveal two crucial observations: (a) the feature domain is better suited than the pixel domain for information compensation due to its stronger spatial and temporal correlations, and (b) warping at an upscaled resolution better preserves high-frequency information, but this benefit is not necessarily monotonic. Therefore, we propose a novel Densely Guided diffusion model with Aligned Features for Video Super-Resolution (DGAF-VSR), with an Optical Guided Warping Module (OGWM) to maintain high-frequency details in the aligned features and a Feature-wise Temporal Condition Module (FTCM) to deliver dense guidance in the feature domain. Extensive experiments on synthetic and real-world datasets demonstrate that DGAF-VSR surpasses state-of-the-art methods in key aspects of VSR, including perceptual quality (35.82\% DISTS reduction), fidelity (0.20 dB PSNR gain), and temporal consistency (30.37\% tLPIPS reduction).
title Rethinking Diffusion Model-Based Video Super-Resolution: Leveraging Dense Guidance from Aligned Features
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.16928