Yang Chen

I am a Ph.D. candidate at Nanyang Technological University, Singapore. Prior to that, I received the M.S. degree at University of Michigan, and obtained B.S. degree at University of Waterloo. My research interests revolve around 3D computer vision and deep learning.

Email  /  Github  /  Google Scholar

Recent Works

We propose Spectral Defense in Gaussian and image fields. Specifically, we design a 3D frequency filter to selectively prune Gaussians exhibiting abnormally high frequencies, and develop a 2D spectral regularization on renderings, distinguishing naturally isotropic frequencies while penalizing anisotropic angular energy to constrain noisy patterns.

Publications

In this paper, we propose MemorySplat, a pose-free 3DGS framework that enforces memory-based temporal coherence for scene rendering and understanding.

We propose view-dependent projection to facilitate point cloud segmentation, designing efficient 3D-to-2D mapping that dynamically adapts to the spatial geometry from view variations. The projections from 3D point distributions produce highly informative single-image inputs, predicting rays inspired by the adaptive behavior of fireworks.

In this paper, we propose Human Gaussian Graph (HGG) to generate generalizable and animatable Gaussian representations. We leverage the human structure prior to recover generalizable and animatable Gaussian representations.

We propose AMContrast3D++ integrating with two branches trained in parallel, where a novel ambiguity prediction module concurrently learns point ambiguities from generated embeddings. We design adaptive objectives for individual points based on their ambiguity levels, aiming to ensure the correctness of low-ambiguity points while allowing mistakes for high-ambiguity points.

In this paper, we introduce the boundary voting network that mitigates feature ambiguity by hierarchically propagating video-level global prior knowledge into local action-transiting regions.

We propose an adaptive margin contrastive learning method for 3D point cloud semantic segmentation, namely AMContrast3D.

Academic Services

• Reviewer, IEEE Transactions on Image Processing (TIP).
• Reviewer, IEEE Transactions on Multimedia (TMM).
• Reviewer, IEEE Transactions on Circuits and Systems for Video (TCSVT).
• Reviewer, IEEE International Conference on Multimedia and Expo (ICME).
• Reviewer, Pattern Recognition (PR).
• Reviewer, Neurocomputing (NEUCOM).
• Reviewer, Chinese Conference on Pattern Recognition and Computer Vision (PRCV).

Honors and Awards

• Academic Achievement on the Dean’s Honors List, University of Waterloo.
• Faculty of Science-Chinese Program Award and Scholarship, University of Waterloo.
• The 7th Blue Bridge Cup Competition, C/C++ Group, The 3rd Prize.
• China CCTV Star of Outlook English Talent Competition, Shandong Division, The 2nd Prize.

Misc

Website Template