Deep learning-based end-to-end depth estimation from a single-frame fringe pattern with the FDSUNet++ network

Xinjun Zhu; Jialei Zhang; Limei Song; Hongyi Wang; Ruiqin Tian

doi:10.1007/s11801-026-4199-9

Optoelectronics Letters ›› 2026, Vol. 22 ›› Issue (6) :373 -378. DOI: 10.1007/s11801-026-4199-9

Article

research-article

Deep learning-based end-to-end depth estimation from a single-frame fringe pattern with the FDSUNet++ network

Author information +

History +

PDF

Abstract

Recently the depth estimation methods based on deep learning (DL) retain challenging to estimate a high-precision depth map in fringe projection structured light three-dimensional (3D) measurement with limited information from a single-frame fringe pattern. In this letter, we proposed a FDSUNet++ convolutional neural network (CNN), which consists of a UNet++ base model, an improved squeeze-and-excitation (ISE) block, a Fourier transform (FT) data preprocessing block, and a discrete wavelet transform (DWT) block. The proposed ISE block can improve the ability of feature extraction and the designed FT data preprocessing block preserves the key features of the fringe pattern by FT. The introduced DWT block reduces the complexity and training cost of the model. By integrating these three blocks into the UNet++, it can better achieve depth estimation. Experimental results from two structured light datasets demonstrate that the proposed FDSUNet++ outperforms the state-of-the-art networks, achieving the best performance in both qualitative and quantitative evaluation.

Keywords

Cite this article

Download citation ▾

Xinjun Zhu, Jialei Zhang, Limei Song, Hongyi Wang, Ruiqin Tian. Deep learning-based end-to-end depth estimation from a single-frame fringe pattern with the FDSUNet++ network. Optoelectronics Letters, 2026, 22 (6) : 373-378 DOI:10.1007/s11801-026-4199-9

登录浏览全文

4963

注册一个新账户忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Marrugo A G, Gao F, Zhang S. State-of-the-art active optical techniques for three-dimensional surface metrology: a review. Optical society of America, 2020, 37(9): B60-B77. J].

[2]	Jason G. Structured-light 3D surface imaging: a tutorial. Advance in optics and photonics, 2011, 3(2): 128-160. J].

[3]	Xu J, Zhang S. Status, challenges and future perspectives of fringe projection profilometry. Optics and lasers in engineering, 2020, 1(135): 106193. J].

[4]	Jiang C, Kilcullen P, Liu X L, et al.. Real-time high-speed three-dimensional surface imaging using band-limited illumination profilometry with a coaxpress interface. Optics letters, 2020, 45(4): 964-967. J].

[5]	Zhang P, Kai Z, Li Z W, et al.. High dynamic range 3D measurement based on structured light: a review. Advanced manufacturing science and technology, 2021, 1(2): 2021004. J].

[6]	Cai Z W, Liu X L, Ped G, et al.. Accurate depth estimation in structured light fields. Optics express, 2019, 27(9): 13532-13546. J].

[7]	Lv S Z, Kemao Q. Modeling the measurement precision of fringe projection profilometry. Light: science and applications, 2023, 12(1): 257. J].

[8]	Nguyen A H, Rees O, Wang Z Y. Learning-based 3D imaging from single structured-light image. Graphical models, 2023, 126: 101171. J].

[9]	Cai Y, Guo M Y, Wang C Y, et al.. TTFDNet: precise depth estimation from single-frame fringe patterns. Sensors, 2024, 24(14): 4733. J].

[10]	Jeught S, Dirckx J. Deep neural networks for single shot structured light profilometry. Optics express, 2019, 27(12): 17091-17101. J].

[11]	Nguyen H, Wang Y Z, Wang Z Y. Single-shot 3D shape reconstruction using structured light and deep convolutional neural networks. Sensors, 2020, 20(13): 3718. J].

[12]	Wang L L, Xue W K, Wang C Y, et al.. Depth estimation from a single-shot fringe pattern based on DD-Inceptionv2-UNet. Applied optics, 2023, 62(34): 9144-9155. J].

[13]	Wang C, Zhou P, Zhu J P. Deep learning-based end-to-end 3D depth recovery from a single-frame fringe pattern with the MSUNet++ network. Optics express, 2023, 31(20): 33287-33298. J].

[14]	Zhu X J, Han Z Q, Song L M, et al.. Wavelet based deep learning for depth estimation from single fringe pattern of fringe projection profilometry. Optoelectronics letters, 2022, 14(18): 699-704. J].

[15]	Yin W, Che Y X, Li X S, et al.. Physics-informed deep learning for fringe pattern analysis. Opto-electron advances, 2024, 7(1): 230034. J].

[16]	Hu J, Shen L, Sun G. Squeeze-and-excitation networks. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 18–23, 2018, Salt Lake City, UT, USA, 2018. New York, IEEE: 7132-7141. [C].

[17]

Zhou Z, Siddiquee MMR, Tajbakhsh N, et al.. UNet++: a nested U-Net architecture for medical image segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, September 20, 2018, Granada, Spain, 2018. New York, IEEE: 3-11. C].

[18]	Xue S K, Qiu W Y, Liu F, et al.. Wavelet-based residual attention network for image super-resolution. Neurocomputing, 2020, 1(382): 116-126. J].

[19]	OLAF R, PHILIPP F, THOMAS B. U-Net: convolutional networks for biomedical image segmentation[EB/OL]. (2015-05-18) [2025-08-25]. https://arxiv.org/abs/1505.04597.

[20]	Fan C M, Liu T J, Liu K H. SUNet: swin transformer UNet for image denoising. 2022 IEEE International Symposium on Circuits and Systems, May 27–June 1, 2022, Austin, TX, USA, 2022. New York, IEEE: 2333-2337. [C].

[21]	Yuan M K, Zhu X J, Hou L P. Depth estimation from single frame fringe projection pattern based on R2U-Net. Laser optoelectronics progress, 2022, 59(16): 83-94. [J].

[22]	Zhu X J, Han Z Q, Zhang Z Z, et al.. PCTNet: depth estimation from single structured light image with a parallel CNN-transformer network. Measurement science and technology, 2023, 34(8): 5402. J].