TENet: Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network

Wang, Ying; Fu, Tao; Zhou, Yu; Kong, Qinglei; Yu, Wenshuai; Liu, Jian; Wang, Yi; Chen, Bo

doi:10.3390/s25030715

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

TENet: Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network

by

Ying Wang

^1,2,

Tao Fu

^1,2,

Yu Zhou

³,

Qinglei Kong

^1,2

,

Wenshuai Yu

⁴

,

Jian Liu

^1,2,*,

Yi Wang

^1,2 and

Bo Chen

^1,2,*

¹

School of Aerospace, Harbin Institute of Technology Shenzhen, Shenzhen 518055, China

²

Key Laboratory of Aerospace RS Big-Data Intelligent Processing and Application of Guangdong Higher Education Institutes, Harbin Institute of Technology Shenzhen, Shenzhen 518055, China

³

Xi’an Institute of Surveying and Mapping, Xian 710054, China

⁴

The College of Civil and Transportation Engineering, Shenzhen University, Shenzhen 518060, China

^*

Authors to whom correspondence should be addressed.

Sensors 2025, 25(3), 715; https://doi.org/10.3390/s25030715

Submission received: 5 December 2024 / Revised: 21 January 2025 / Accepted: 23 January 2025 / Published: 24 January 2025

(This article belongs to the Section Sensor Networks)

Download Versions Notes

Abstract

Oblique photogrammetry imagery often suffers from uneven resolution and blurred details, leading to poor surface texture quality in 3D reconstructions, particularly for building facades. To address these challenges, we propose a novel Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network (TENet) and introduce a comprehensive 3D texture enhancement pipeline. This pipeline applies 2D texture super-resolution techniques to 3D models for fine-grained texture restoration, enhancing the surface texture quality. TENet leverages attention mechanisms and frequency-domain techniques to improve texture sharpness and edge accuracy. Our approach includes a Region-Resolution Adaptive Enhancement Module (RAEM) and a Frequency-Domain Edge Enhancement Mechanism (FDEEM) to enhance local details and restore critical edge features. The experimental results demonstrate that TENet outperforms existing methods, significantly improving texture quality and 3D reconstruction performance. Ablation studies confirmed the effectiveness of each component in enhancing 3D texture reconstruction. The network is validated for real-world applications, showing its ability to significantly reduce edge artifacts and restore clear, accurate textures in real-world 3D surface models.

Keywords: 3D surface texture enhancement; oblique photogrammetry; self-attention mechanism; frequency domain enhancement

Share and Cite

MDPI and ACS Style

Wang, Y.; Fu, T.; Zhou, Y.; Kong, Q.; Yu, W.; Liu, J.; Wang, Y.; Chen, B. TENet: Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network. Sensors 2025, 25, 715. https://doi.org/10.3390/s25030715

AMA Style

Wang Y, Fu T, Zhou Y, Kong Q, Yu W, Liu J, Wang Y, Chen B. TENet: Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network. Sensors. 2025; 25(3):715. https://doi.org/10.3390/s25030715

Chicago/Turabian Style

Wang, Ying, Tao Fu, Yu Zhou, Qinglei Kong, Wenshuai Yu, Jian Liu, Yi Wang, and Bo Chen. 2025. "TENet: Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network" Sensors 25, no. 3: 715. https://doi.org/10.3390/s25030715

APA Style

Wang, Y., Fu, T., Zhou, Y., Kong, Q., Yu, W., Liu, J., Wang, Y., & Chen, B. (2025). TENet: Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network. Sensors, 25(3), 715. https://doi.org/10.3390/s25030715

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

TENet: Attention-Frequency Edge-Enhanced 3D Texture Enhancement Network

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI