EvTexture++: Event-Driven Texture Enhancement for Video Super-Resolution

2026-06-11

AI Summary

Existing video super-resolution (VSR) methods struggle to recover fine textures and maintain temporal consistency in scenes with large motion, and they fail to effectively exploit the high temporal resolution of event camera data. To address this, this work proposes EvTexture++, the first framework specifically designed to leverage event data for enhancing VSR texture details. Its core components include a plug-and-play iterative texture enhancement module and an event-guided temporal texture alignment mechanism, which uniquely employs event signals for texture enhancement rather than solely for motion compensation. Evaluated on five benchmark datasets, EvTexture++ achieves state-of-the-art performance; when integrated into existing VSR models, it yields up to a 1.55 dB PSNR improvement on Vid4.

Abstract

Event-based vision has drawn increasing attention owing to its distinctive properties, including ultra-high temporal resolution and extreme dynamic range. Recent works have introduced it to video super-resolution (VSR) to enhance flow estimation and temporal alignment. In contrast, this paper shifts the focus of event signals from motion refinement to texture enhancement in VSR. We propose EvTexture++, the first event-driven framework dedicated to texture enhancement in VSR. It leverages high-frequency spatiotemporal details from events to improve texture recovery. EvTexture++ incorporates a customized texture enhancement branch, along with an iterative texture enhancement module that progressively exploits high-temporal-resolution event information for texture restoration. This enables gradual refinement of texture regions across iterations, yielding more accurate and detailed high-resolution outputs. Besides intra-frame texture recovery, large motions could degrade inter-frame temporal consistency, particularly in texture regions, leading to texture flickering. To mitigate this, we further exploit the continuous-time motion cues of events to enhance temporal consistency, introducing a temporal texture alignment module that estimates event-guided texture-aware flow for precise inter-frame texture alignment. Moreover, EvTexture++ is designed as a plug-and-play tool to flexibly boost the performance of existing VSR models. Experiments on five datasets demonstrate that EvTexture++ achieves state-of-the-art performance. When integrated into recent VSR models, it yields significant improvements, with gains of up to 1.55 dB in PSNR on the texture-rich Vid4 dataset. Code: https://github.com/DachunKai/EvTexture.

Problem

• video super-resolution
• texture enhancement
• temporal consistency
• event-based vision
• texture flickering

Innovation

• event-based vision
• video super-resolution
• texture enhancement
• temporal consistency
• plug-and-play framework