About Me
I am a PhD student at Simon Fraser University, advised by Prof. Jiangchuan (JC) LIU. I received my Master’s degree at Southeast University, advised by Prof. Fang Dong, and my Bachelor’s degree from Jiangnan University in 2020, advised by Prof. Ya Guo. I design practical systems for photorealistic volumetric video reconstruction and transmission, targeting downstream applications in 3D scene understanding, embodied intelligence, and immersive volumetric telepresence.
Research Interest
Education Experience
- Simon Fraser University, Burnaby, Canada
- Ph.D. in Computing Science.
- August 2023 -
- --------------------------------
- Southeast University, Nanjing, China
- M.Eng. in Computer Science and Engineering.
- August 2020 - July 2023
- --------------------------------
- Jiangnan University, Wuxi, China
- August 2016 - July 2020
- B.Eng. in Internet of Things (IoT) Engineering
- GPA: 3.63, rank 9/141.
Publications
Daheng Yin, Yili Jin, Jianxin Shi, Isaac Ding, Miao Zhang, Fangxin Wang, Zhaowu Huang, Cong Zhang, Jiangchuan Liu, Fang Dong “CAGS: Color-Adaptive Volumetric Video Streaming with Dynamic 3D Gaussian Splatting,” SIGGRAPH 2026 Conference Papers (SIGGRAPH’26), 2026, GitHub repo
Daheng Yin, Isaac Ding, Yili Jin, Jianxin Shi, Jiangchuan Liu “TrackerSplat: Exploiting Point Tracking for Fast and Robust Dynamic 3D Gaussians Reconstruction,” SIGGRAPH Asia 2025 Conference Papers (SA’25), 2025, GitHub repo
Zhaowu Huang, Fang Dong, Xiaolin Guo, Daheng Yin “FaSei: Fast Serverless Edge Inference with Synergistic Lazy Loading and Layer-wise Caching,” IEEE International Conference on Computer Communications (INFOCOM), 2025
Daheng Yin, Jianxin Shi, Miao Zhang, Zhaowu Huang, Jiangchuan Liu, Fang Dong “FSVFG: Towards Immersive Full-Scene Volumetric Video Streaming with Adaptive Feature Grid,” 32nd ACM International Conference on Multimedia (MM’24), 2024
Daheng Yin, Fang Dong, Baijun Chen, Dian Shen, Ruiting Zhou, Xiaolin Guo, Zhaowu Huang “WAEVSR: Enabling Collaborative Live Video Super-Resolution in Wide-Area MEC Environment,” IEEE/ACM 31st International Symposium on Quality of Service (IWQoS), 2023
Baijun Chen, Daheng Yin, Lifei Teng, Fang Dong “HyperRTV: Neural-Enhanced Adaptive Real-Time Video Streaming Based on Terminal-Edge Collaboration,” 26th International Conference on Computer Supported Cooperative Work in Design (CSCWD), 2023
Xiaolin Guo, Fang Dong, Dian Shen, Zhaowu Huang, Zhenyang Ni, Yulong Jiang, Daheng Yin “Exploiting the computational path diversity with in-network computing for MEC,” 19th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON), 2022
Mengyang Liu, Anran Tang, Huitian Wang, Lin Shen, Yunhan Chang, Guangxing Cai, Daheng Yin, Fang Dong, Wei Zhao “Accelerating Multi-Object Tracking in Edge Computing Environment with Time-Spatial Optimization,” Ninth International Conference on Advanced Cloud and Big Data (CBD), 2021
Open-source Projects
gaussian-splatting 
Refactored Python training and inference package for 3D Gaussian Splatting, preserving the original algorithms while reorganizing the code into a standard Python package. The project adds practical training features including exposure compensation, camera and 3DGS joint optimization, depth regularization, local relative depth regularization, image masks, and integrated gsplat / 2DGS backends.
InstantSplat 
Refactored Python package for InstantSplat, supporting sparse-view Gaussian Splatting initialization and camera/3DGS joint training. It includes DUSt3R, MASt3R, COLMAP sparse/dense reconstruction, VGGT, and Map-Anything based initialization workflows.
reduced-3dgs 
Refactored Python package for Reduced-3DGS, maintaining the original memory-footprint-reduction algorithms while making the code easier to install and reuse. It provides pruning, SH culling, and K-Means vector quantization for compact 3D Gaussian Splatting models, and builds on the packaged gaussian-splatting project above.
feature-3dgs 
Refactored Python package for Feature 3DGS, adding a modular Extractor-Decoder architecture for semantic feature distillation on Gaussian points. It supports DINOv3-based dense features, PCA visualization, interactive viewing, and the training modes inherited from packaged gaussian-splatting.
3dgs-mcmc 
Refactored Python package for 3D Gaussian Splatting as Markov Chain Monte Carlo, providing an installable MCMC trainer for 3D Gaussian Splatting and integration with reduced-3dgs.
gscompressor 
Compresses 3DGS scenes with Draco and works with gaussian-splatting, reduced-3dgs, and lapis-gs.
lapis-gs 
Refactored Python package for LapisGS, a layered progressive 3D Gaussian Splatting method for adaptive streaming. It provides multi-resolution training, progressive layer extraction, and integration with gaussian-splatting and reduced-3dgs.
feature-4dgs 
Sequence-aware training extension for Feature 3DGS, built on top of feature-3dgs and gaussian-splatting. It extends semantic feature distillation from single scenes to multi-timestep 4D / dynamic-scene training with per-frame Gaussian models and one shared decoder for a consistent feature space.
track-4dgs 
Point tracking extension for 4D Gaussian Splatting. It wraps sequence point trackers such as CoTracker3 and VGGT, supports single-view image-sequence tracking, and projects/tracks 3D Gaussians across multi-timestep Gaussian Splatting camera datasets.
Contest
- SIGGRAPH Asia 2025 Volumetric Video Workshop, Compression Track 3rd, 2025.08 - 2025.12
- TensorRT Hackathon 2022, Winner Prize, by NVIDIA & Alibaba Cloud TIANCHI, 2022.03 - 2022.5
- TensorRT Hackathon 2021, Ranking 4/48, by NVIDIA & Alibaba Cloud TIANCHI, 2021.03 - 2021.5
- National College Mathematical Contest in Modeling, 2nd Prize (National), 2018.09
- 9th National College Mathematical Contest, 2nd Prize (Provincial), 2017.11
- 14th Jiangsu College Mathematical Contest, 1st Prize, 2017.05
Honors & Awards
- Outstanding Graduate of Jiangnan University, 2020.06
- China National Scholarship (2016-2017), 2017.11