Yaoxing Wang

I am currently a Ph.D. student at the Unmanned Systems Technology Research Institute of Northwestern Polytechnical University, supervised by Ass. Prof. Shan Gao. Prior to that, I received my Master of Electronic Information degree in computer technology from the School of Information Engineering at Ningxia University in 2024, supervised by Prof. Hao Liu. In 2022, I received my Bachelor of Engineering degree in computer science and technology at Ningxia University, Yinchuan, China.

My research interests mainly include computer vision, 3D reconstruction, and thermal infrared reconstruction.

Email  /  Google Scholar  /  OpenReview  / 

A frequency-gated graph splatting framework that disentangles diffusive low-frequency thermal distributions from sparse high-frequency structural boundaries for high-fidelity thermal novel view synthesis.

A physics-aware framework for generating realistic full-body human motion that interacts with multiple objects, incorporating physical constraints and dynamics for natural movement synthesis.

A physics-inspired framework that couples adaptive spatial support with internal spectral modulation via a unified generalized-exponential Gabor primitive for high-fidelity thermal infrared novel view synthesis.

A vision–language framework that aligns hierarchical textual affiliations with multiscale visual features via optimal transport, addressing scale variation and semantic ambiguity in whole-body pose estimation.

A SAM-based framework incorporating generative latent space enhancement via pre-trained diffusion models, enabling robust segmentation across images of varying quality and degradation levels.

A self-supervised learning framework with structural equivariance for facial pose estimation.

The proposed PML typically incorporates with the ordinal margin and the variational margin, simultaneously plugging in the globally-tuned deep neural network paradigm.

^† Co-first author   |   ^✉ Corresponding author

Website Template