~*~ Welcome to ~*~
Ben Huang's Homepage

Converts real-world scenes into Minecraft worlds using computer vision and neural radiance fields.

Worldcraft processes photographs through a four-stage pipeline — semantic segmentation with Mask2Former, NeRF training via Nerfstudio, point cloud voxelization, and Minecraft block conversion — to create playable 3D worlds from real-life images.

High-performance CUDA library for graph algorithms and processing.

This project demonstrates your skills in front-end development and showcases your ability to create responsive, user-friendly web applications. It includes modern design principles, accessibility features, and cross-browser compatibility.

Full‑stack MERN app with responsive UI and a robust Node/Express API.

This is a description of your second project. Click to expand and see more details about your work.

This project showcases your full-stack development capabilities and demonstrates your understanding of modern web technologies. It features a responsive frontend built with React, a robust backend API, and efficient database management.

Data driven Reinforcement Learning application to provide intelligent hashrate predictions

This is a description of your third project. Click to expand and see more details about your work.

This project demonstrates your expertise in Python web development and database design. It includes advanced features like user authentication, data visualization, and RESTful API endpoints.

Scan your home appliances with on-device AI to monitor energy usage and detect ghost energy drain.

Built at SF Hacks 2026, WattVision uses a React Native + Expo frontend with a Python FastAPI backend and on-device machine learning to identify power-hungry appliances across web, Android, and iOS.