Rong Liu

AtomGS: Atomizing Gaussian Splatting for High-Fidelity Radiance Field

Rong Liu, Rui Xu, Yue Hu, Meida Chen, Andrew Feng

Pre-Print arXiv:2405.12369

Project Page | Paper | Code

Universal Photorealistic Style Transfer: A Lightweight and Adaptive Approach

Rong Liu, Enyu Zhao, Zhiyuan Liu, Andrew Feng, Scott John Easley

Pre-Print arXiv:2309.10011

Project Page | Paper | Code

Gaussian Splatting Monitor

We aim to extend 3D Gaussian Splatting, currently limited to RGB rendering and loss, by incorporating differentiable outputs inspired by the versatile Neural Radiance Fields (NeRF), thus enabling advanced research and novel loss functions for improved volumetric analysis.

Project Page | Code

GaussianNeRF: High-Fidelity Surface Reconstruction based on Gaussian Constraint

We enhanced NeRF’s mesh modeling quality and speed by incorporating a Gaussian loss function, resulting in better mesh quality and outperforming Nerfacto in texture and runtime.

Report

GPT-ID: GPT Interaction Detector

We propose GPT-ID, a tool built upon DetectGPT, designed to distinguish AI-generated from human-written texts with improved accuracy and novel edit distribution comparison, aiming to ensure responsible use of text-generation technologies.

Report

Store Sales - Time Series Forecasting

The project aims to forecast sales for Ecuador’s Corporación Favorita using time series data, achieving top 5 leaderboard ranking through data analysis, Exponential Smoothing, LightGBM, and Genetic Mutation algorithms.

Kaggle Competition | Report