Yucheng Zheng

ORCID: 0000-0002-0779-9703
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Vision and Imaging
  • Image Processing Techniques and Applications
  • Random lasers and scattering media
  • Optical measurement and interference techniques
  • Digital Holography and Microscopy
  • Advanced Optical Sensing Technologies
  • Advanced Image Processing Techniques
  • Face recognition and analysis
  • Optical Coherence Tomography Applications
  • Facial Rejuvenation and Surgery Techniques
  • Generative Adversarial Networks and Image Synthesis
  • Facial Nerve Paralysis Treatment and Research

Shanghai Jiao Tong University
2024

University of California, Riverside
2019-2023

Mask-based lensless cameras replace the lens of a conventional camera with custom mask. These can potentially be very thin and even flexible. Recently, it has been demonstrated that such mask-based recover light intensity depth information scene. Existing recovery algorithms either assume scene consists small number planes or solve sparse problem over large 3D volume. Both these approaches fail to scenes variations. In this paper, we propose new approach for estimation based on an...

10.1109/tci.2020.3010360 article EN IEEE Transactions on Computational Imaging 2020-01-01

Lensless cameras provide a framework to build thin imaging systems by replacing the lens in conventional camera with an amplitude or phase mask near sensor. Existing methods for lensless can recover depth and intensity of scene, but they require solving computationally-expensive inverse problems. Furthermore, existing struggle dense scenes large variations. In this paper, we propose system that captures small number measurements using different patterns on programmable mask. context, make...

10.1109/iccv48922.2021.00260 article EN 2021 IEEE/CVF International Conference on Computer Vision (ICCV) 2021-10-01

Mask-based lensless cameras can be flat, thin, and light-weight, which makes them suitable for novel designs of computational imaging systems with large surface areas arbitrary shapes. Despite recent progress in cameras, the quality images recovered from is often poor due to ill-conditioning underlying measurement system. In this paper, we propose use coded illumination improve reconstructed cameras. our model, scene/object illuminated by multiple patterns as camera records sensor...

10.1109/tci.2023.3234898 article EN publisher-specific-oa IEEE Transactions on Computational Imaging 2023-01-01

Mask-based lensless cameras replace the lens by placing a fixed mask on top of an image sensor. These can potentially be very thin and even flexible. Recently, it has been demonstrated that such mask-based recover light intensity depth information scene. Existing recovery algorithms either assume scene consists small number planes or solve sparse problem over large 3D volume, lose robustness to complicated scenes consisting varying surface. In this paper, we propose new approach for...

10.1109/camsap45676.2019.9022507 article EN 2019-12-01

Contemporary makeup approaches primarily hinge on unpaired learning paradigms, yet they grapple with the challenges of inaccurate supervision (e.g., face misalignment) and sophisticated facial prompts (including parsing, landmark detection). These prohibit low-cost deployment models, especially mobile devices. To solve above problems, we propose a brand-new paradigm, termed "Data Amplify Learning (DAL)," alongside compact model named "TinyBeauty." The core idea DAL lies in employing...

10.48550/arxiv.2403.15033 preprint EN arXiv (Cornell University) 2024-03-22

Recently, mask-based camera has been proposed by replacing the lens with a coded-mask that modulates light as it reaches sensor. Such cameras can reliably capture both direction and depth information about scene. However, current designs of lensless contain only one sensor array. In contrast to that, using multiple distributed arrays image scene may offer us many advantages, such more flexible placing strategy different views this paper, we present line cameras, in which replace rectangular...

10.1109/ieeeconf44664.2019.9048751 article EN 2019-11-01

Mask-based lensless cameras offer an alternative option to conventional cameras. Compared cameras, can be extremely thin, flexible, and lightweight. Despite these advantages, the quality of images recovered from is often poor because ill-conditioning underlying linear system. In this paper, we propose a new method address problem illconditioning by combining coded illumination patterns with mask-based imaging. We assume that object illuminated multiple binary camera acquires sequence for...

10.1109/icassp40776.2020.9052955 article EN ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2020-04-09

Mask-based lensless cameras offer a novel design for imaging systems by replacing the lens in conventional camera with layer of coded mask. Each pixel encodes information entire 3D scene. Existing methods reconstruction from measurements suffer poor spatial and depth resolution. This is partially due to system ill conditioning that arises because point-spread functions (PSFs) different planes are very similar. In this paper, we propose capture multiple scene under sequence illumination...

10.1109/ojsp.2022.3231180 article EN cc-by IEEE Open Journal of Signal Processing 2022-01-01

Mask-based lensless cameras can be flat, thin, and light-weight, which makes them suitable for novel designs of computational imaging systems with large surface areas arbitrary shapes. Despite recent progress in cameras, the quality images recovered from is often poor due to ill-conditioning underlying measurement system. In this paper, we propose use coded illumination improve reconstructed cameras. our model, scene/object illuminated by multiple patterns as camera records sensor...

10.48550/arxiv.2111.12862 preprint EN other-oa arXiv (Cornell University) 2021-01-01

Mask-based lensless cameras replace the lens of a conventional camera with custom mask. These can potentially be very thin and even flexible. Recently, it has been demonstrated that such mask-based recover light intensity depth information scene. Existing recovery algorithms either assume scene consists small number planes or solve sparse problem over large 3D volume. Both these approaches fail to scenes variations. In this paper, we propose new approach for estimation based on an...

10.48550/arxiv.1910.02526 preprint EN other-oa arXiv (Cornell University) 2019-01-01

Lensless cameras provide a framework to build thin imaging systems by replacing the lens in conventional camera with an amplitude or phase mask near sensor. Existing methods for lensless can recover depth and intensity of scene, but they require solving computationally-expensive inverse problems. Furthermore, existing struggle dense scenes large variations. In this paper, we propose system that captures small number measurements using different patterns on programmable mask. context, make...

10.48550/arxiv.2108.07966 preprint EN other-oa arXiv (Cornell University) 2021-01-01
Coming Soon ...