Photo

Patt PHURTIVILAI

Hello! I’m Patt Phurtivilai, a research postgraduate in Artificial Intelligence at the University of Hong Kong. I conduct 3D vision research in CGVU Lab under the supervision of Prof. Taku Komura.

I am passionate about building AI that understands and interacts with the real 3D world. My research focuses on computer vision and graphics, including 3D tracking, reconstruction, and neural rendering.

My current work explores self-supervised multi-view tracking and 3D perception systems developed through collaborations with academic and industry researchers.

   /      /   pattgene [at] connect (dot) hku (dot) hk



Academic Publications

TrackFish3D method overview
TrackFish3D: Self-Supervised 3D Tracking of Schooling Fish from Multi-view Videos
Patt Phurtivilai, Zhiyang Dou,† Yifan Wu, Kinfung Chu, Yuan Liu, Lei Yang, Wenping Wang, Taku Komura†
Accepted at NeurIPS 2026

TL;DR: A self-supervised multi-view 3D tracking method that learns fish identity consistency from geometry alone, without identity labels or appearance cues.

[Project Page] [Paper (Coming Soon)] [Code (Coming Soon)]
EndoX 3D Gaussian Splatting results
EndoX: A GPU-Accelerated Endoscopic Perception Simulation Framework
Lam Kwan Tung, Ho Ping Shu, Patt Phurtivilai, Cheung Ka Chun
Accepted at MICCAI 2026

TL;DR: A GPU-accelerated simulation framework for endoscopic perception that supports fast, realistic experimentation for 3D scene understanding and rendering research.

[MICCAI Page] [Project Page] [Paper (Coming Soon)] [Code]
DensifyBeforehand: LiDAR-assisted Content-aware Densification for Efficient and Quality 3D Gaussian Splatting
Patt Phurtivilai, Huang Leyang, Yinqiang Zhang, Lei Yang
CVM 2026 - Technical Brief

TL;DR: Densify 3DGS before training using sparse LiDAR + RGB monocular depth and ROI-aware sampling, avoiding wasteful adaptive densification. Achieves similar quality to SOTA with less compute/training time.

[Page] [Paper]
3D Digital Twins for MakerSpace
Patt Phurtivilai, Lio Qing, Ryan Chan
Presented at 9th International Symposium on Academic Makerspaces (ISAM2025)

[ISAM Page]

Fostering AI Makerspace: A Journey through Introductory Hands-On Workshop
Leyang Huang, Patt Phurtivilai, Xiang Chen, Syed Nafis Ishrak, Jose Alberto Espino Pitti
Presented at 8th International Symposium on Academic Makerspaces (ISAM2024)

Recent Projects

InnoGrow: Non-Toxic Weed Killing Box

  • Developed INNOGROW: a non-toxic laser weed control solution for sustainable farming in Asia.
  • Trained a weed detection model with a custom dataset; integrated with NCNN on Raspberry Pi for a compact design.
  • Controlled laser shooting with a servo using an ESP32 board.
[Video] [Code]
Digital Twins: 3D Gaussain Splatting Reconstruction

  • Collaborating with MTR, building a digital twins for tunnels.
  • Lidar Scaning and processing.
  • 3D Gaussian Splatting using Lidar Slam initialization.
Multimodal AI Assistant: RAG system for both image and text

  • Developed a multimodal RAG system for processing and generating text and images.
  • Created a data preprocessing pipeline with layout detection and Azure OCR for content extraction.
  • Designed a retrieval mechanism for interleaving text and images, enhancing traceability and accessibility.
[Video] [Code]
Grounding-Dino: zero shot training with real time camrea feed
participate in HKU innoshow2024

  • Developed a real-time object detection and tracking system from Grounding DINO and live feed.
  • Implemented keyword-based detection for continuous object tracking.
  • Showcased advanced computer vision techniques for dynamic mobile interaction.
[Video] [Code]
Azai Solution - CRM + Payment API Application
Winning the 1st prize for Cyberport CUPP 2023 micro funding

  • Developed AZAI, a smart CRM app for landlords in Thailand using React Native and TypeScript.
  • Integrated payment APIs and Firebase for efficient contract and property management.
  • Utilized Next.js for backend development, adding analytics features to improve landlord-resident interactions.
[Page] [Video]


Experience

Research Assistant @ CGVU Lab, HKU


  • Conducting independent research on an Object Tracking System, focusing on multi-camera association and effective triangulation techniques
  • Generating 3D synthetic motion data using Unreal Engine 5.

NVIDIA AI Research Collaborator


  • Co-authored two research papers with NVIDIA AI researchers by developing depth estimation and 3D Gaussian Splatting models for endoscopic perception.

Summer AI Research Assistant @ Innowing, HKU


  • Conducted individual research in computer graphics and 3D reconstruction, focusing on Gaussian Splatting for real-time rendering using Lidar, depth sensors, and SLAM technologies.
  • Developed a Multimodal Retrieval-Augmented Generation system by researching LLMs, OCR, and layout detection models, utilizing GPT-4.
  • Collaborated with UBC and Tsinghua University on research projects and hosted workshops on computer graphics, computer vision, and retrieval-augmented generation.

Part-time AI Research Assistan @ Innowing, HKU


  • Computer vision and graphics research team, exploring cutting-edge technologies in AI. Collaboration on a real work with MTR corporation, scaning tunnel and create digital twins for them.
  • Hosted workshops on computer vision, computer graphics, and large language models for participants from Hong Kong University.
  • Conducted research on large language models and retrieval-augmented generation (RAG), developing a local chatbot for university.

Founder and CEO of AZAI Solution, a self-startup funded by Cyberport.


  • Founded and served as CEO of AZAI, a fintech startup awarded a 100k HKD grant from Cyberport for innovative solutions.
  • Developed an Android application and website to help landlords in Thailand create contracts, manage payments, and oversee property management.
  • Designed AZAI as a CRM application utilizing payment APIs, providing data analytics services to facilitate interactions between landlords and residents.

Software Engineering & Data Analyst Summer Internship @ Risksis Technology Limited


  • Engaged in data retrieval, preprocessing, and model training using Keras, OpenCV, and Scikit-learn during a summer internship.
  • Developed a web application utilizing frontend frameworks, including an interface for 3D visualization with Unity.
  • Wrote scripts in C# and Python to meet project requirements and enhance functionality.


Teaching / Workshop

2024
AI-powered Object Detection workshop
AI-powered Object Detection
HKU

Theory and hands-on with YOLO and GroundingDINO.

[Workshop Page]
AIHA workshop
AIHA (AI Historian Assistant): RAG Workshop
HKU

Theory and hands-on for an end-to-end multimodal RAG system.

[Workshop Page]
2025
Intro to PyTorch workshop
Intro to PyTorch: internal workshop for SRA interns
HKU

Theory and hands-on for PyTorch.

How We Turned Ideas into Innovation workshop
How We Turned Ideas into Innovation
HKU [Workshop Page]