Jun Zhao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Jun Zhao has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Chen Dai
University of Science and Technology of China
2 shared publications - Feihu Xu
University of Science and Technology of China
2 shared publications - Jianwei Zeng
University of Science and Technology of China
2 shared publications - Ruilin Ye
University of Science and Technology of China
2 shared publications - Wenqing Hong
Kunming Institute of Physics
2 shared publications - Wenwen Li
University of Science and Technology of China
2 shared publications
Latest Publications
- Open Access
- 25 August 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1
Erratum: Thermal Non-Line-of-Sight Imaging through Rough Surfaces
This is an erratum for the article “Thermal Non-Line-of-Sight Imaging through Rough Surfaces” published in ACM Trans. Graph. 45, 5, Article 41 (June 2026), 21 pages.
- 25 August 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1
Open AccessErratum: Thermal Non-Line-of-Sight Imaging through Rough Surfaces
This is an erratum for the article “Thermal Non-Line-of-Sight Imaging through Rough Surfaces” published in ACM Trans. Graph. 45, 5, Article 41 (June 2026), 21 pages.
- Open Access
- 29 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1-21
Thermal Non-Line-of-Sight Imaging through Rough Surfaces
Thermal non-line-of-sight (NLOS) imaging utilizes the thermal radiation emitted by hidden objects to “see around corners”, offering promising applications in security and search-and-rescue missions. However, reconstructing hidden scenes through rough relay walls remains highly challenging due to severe spatial mixing caused by complex reflection and scattering. While existing model-based and learning-based approaches yield high-quality results under reflective surfaces, they struggle to deliver satisfactory performance under rough surfaces. Here we introduce NLOSFormer, a physics-embedded neural network to address this challenge. By formulating light transport as a convolution process, NLOSFormer estimates the convolution kernel from measurements to guide reconstruction within an end-to-end network. This explicitly estimated kernel effectively narrows the solution space, significantly improving generalization and enabling relative depth estimation. To support training and evaluation, we further present a publicly available dataset named ThermalNLOS, together with a novel data augmentation strategy to alleviate overfitting. Extensive experiments demonstrate that NLOSFormer significantly outperforms existing approaches across a wide range of relay walls. Notably, it achieves real-time imaging of dynamic targets under rough surfaces, representing a substantial advance toward robust and practical deployment.
- 29 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (5) , 1-21
Open AccessThermal Non-Line-of-Sight Imaging through Rough Surfaces
Thermal non-line-of-sight (NLOS) imaging utilizes the thermal radiation emitted by hidden objects to “see around corners”, offering promising applications in security and search-and-rescue missions. However, reconstructing hidden scenes through rough relay walls remains highly challenging due to severe spatial mixing caused by complex reflection and scattering. While existing model-based and learning-based approaches yield high-quality results under reflective surfaces, they struggle to deliver satisfactory performance under rough surfaces. Here we introduce NLOSFormer, a physics-embedded neural network to address this challenge. By formulating light transport as a convolution process, NLOSFormer estimates the convolution kernel from measurements to guide reconstruction within an end-to-end network. This explicitly estimated kernel effectively narrows the solution space, significantly improving generalization and enabling relative depth estimation. To support training and evaluation, we further present a publicly available dataset named ThermalNLOS, together with a novel data augmentation strategy to alleviate overfitting. Extensive experiments demonstrate that NLOSFormer significantly outperforms existing approaches across a wide range of relay walls. Notably, it achieves real-time imaging of dynamic targets under rough surfaces, representing a substantial advance toward robust and practical deployment.