Heejun Park's Projects
Uses python Flask server, sqlite, html to create a simple game.
BEVDet λ
Όλ¬Έ 리뷰 λ° μ½λ μ€μ΅
BEVFormer Paper Review and Code Practice
PyTorch BEVFusion Code Practice
μ½ν
λ¬Έμ νμ΄
건κ°κ²μ§κ²°κ³Όν΅λ³΄μλ§ κ°κ³ λ―Έλμ 걸릴 μ μλ μ§λ³μ λν μμΈ‘μ ν΄μ£Όλ 'λ₯ν° AI'
Investigating the Effects of Bag of Tricks for Object Classification Models: ResNet and ViT
2023-2 μ€νμμ€ κΈ°μ¬ μ€μ΅μ μν μ μ₯μμ
λλ€.
Understanding how Layer Normalization works using PyTorch!
Learning the Basics of Linux System
Utility to convert the NYU Depth V2 dataset into point clouds for advanced 3D visualization and analysis.
GitHub Profile ReadMe!
Web-based 3D Point Cloud Visualization and Classification Service. Multi-container Web Application with Docker.
PointBeV is a SOTA sparse segmentation model for BeV perception. I added extensive comments for clarity and optimized some of the original code for better efficiency. I hope this repository helps those who study BeV perception for AVs.
PointNet and PointNet++ implemented by pytorch (pure python) and on ModelNet, ShapeNet and S3DIS.
Performing Semantic Segmentation using PointTransformerV3
CIFAR10 classifier using PyTorch. The classifier uses a CNN.
MNIST digit classifier using PyTorch. The classifier uses a FCN rather than a CNN.
π We developed a Remote Control Autonomous Taxi service called "R-Mutax".
PyTorch version of SCPNet. Clonned from the official repo
Generate Fused MultiFrame semantic-KITTI Dataset
Enhanced version of the official Simple-BEV implementation, with comprehensive comments added to the code for better understanding and readability.
TIL μ€ν°λ
PyTorch implementation of TabNet paper : https://arxiv.org/pdf/1908.07442.pdf
I created a POS tagging LSTM using TF as part of the AI course's assignment task.
Investigating the Vulnerability of 3D Point Cloud Classifiers to Backdoor Attacks Leveraging Transfer Learning
A Study on Transfer Learning-Based Investigation of Backdoor Attack Vulnerabilities in 3D Point Cloud Classifiers
Implementation of Vision Transformer(ViT) using PyTorch. Also using Tensorboard to visualize the results.