Junyu Liu's Projects
Coursework 1 for (COMP2048 UNNC) Algorithm Correctness and Efficiency module
Coursework 2 for (COMP2048 UNNC) Algorithm Correctness and Efficiency
Implementation of As Rigid As Possible paper in an interactive way, using C++, Qt, and OpenGL framework.
Repository of COMP2051-AIM-coursework
BROOK Dataset, a multi-modal and facial video dataset for data-driven Human-Vehicle Interaction.
An Agile RISC-V SoC Design Framework with in-order cores, out-of-order cores, accelerators, and more
CLIPScore EMNLP code
FEM simulator written in C++, Qt, and OpenGL framewrok, supporting 3 different orders integrators, collision resolving, adaptive time stepping, and parallelization
Coursework for COMP2049 Language and Computation
Lecture notes, exercises, homeworks for Brown CS1951x 2023
Coursework repository for COMP3065-CV
This code completes a tutorial about gaussian mixture models (gmm) in python using scikit-learn
Coursework for COMP2047 Introduction to Image Processing
2D interactive drawing board and image editor. The drawing board supports flexible brush flows (constant, linear, quadratic, smudge, spray, fill, and alpha blending), brush sizes, brush colors. The image editor supports image filtering (3 blurring filters, 2 edge detection), image scaling and roatation, tone mapping, etc.
Personal website of Junyu Liu
Coursework for COMP3055 Machine Learning Module
One repository is all that is necessary for Multi-agent Reinforcement Learning (MARL)
Low-level manipulations of triangular meshes by Loop Subdivision, Quadric Error Simplification, Isotropic Remeshing, Bilateral Mesh Denoising.
Synthesized models for PHOG to make the results reproducible by the research community
Physically-based renderer using Monte Carlo sampling of path tracing, which supports 6 BRDFs, soft shadows, indirect illumination, Russian roulette path termination, SDR/HDR tone mapping, attenuate refraction, stratified sampling, depth of field, low discrepancy sampling, BRDF importance sampling, etc.
[CVPR 2022] Pre-Training 3D Point Cloud Transformers with Masked Point Modeling
Classify 3D point clouds with self-generated dataset. Brown Vision Comuting (BVC) starter project for onboard.
PointNet and PointNet++ implemented by pytorch (pure python) and on ModelNet, ShapeNet and S3DIS.
pytorch implementation for "PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation" https://arxiv.org/abs/1612.00593
A polyvalent C++ FEM library
Coursework for (COMP2046 UNNC) Operating System and Concurrency
Ray Tracer which supports 4 BRDFs, implict/explicit shape rendering, advanced lighting, texture mapping,anti-aliasing, super-sampling, filtering, depth of field, soft shadow, acceleration, parallelization, etc.
Realtime simulation for nature, which includes particle systems for snow simulation, procedural generation algorithms for terrain, physics-based dynamic systems for snow-terrain interaction, and a dynamic weather system.