-
YOLO Demo on Static Webpage
Attempted to deploy YOLO to a static web page. Deployed YOLOv8n in ONNX format for easier cross-platform use, executed model inference through WebAssembly, and used Canvas to draw…
-
R2 Gateway: Cost-Controlled S3-Compatible Object Storage Gateway
GITHUB Since my VPS has limited storage, I recently explored object storage solutions. I found that Cloudflare R2 offers a free tier, but it doesn’t provide a reliable way to automatically…
-
AWS Cloud Album Project
Technology Stack Python, Flask, AWS (EC2, S3, SQS, Lambda), SQL, YOLOv8 (ONNX), RESTful API, JWT, OpenStreetMap/Leaflet, HTML/CSS, Docker Project Overview Designed and implemented…
-
Image Compression May Improve Classification Networks’ Accuracy
Python, PyTorch, NumPy, OpenCV, Scikit-learn, Matplotlib, Knowledge Distillation, Deep Learning, Data Preprocessing, Model Evaluation “Figure 9 shows the accuracy …
-
Knowledge Distillation with ResNet for Raw Image Classification
Recently, I conducted a simple experiment on a raw image classification task, comparing the accuracy between three setups: baseline without augmentation, ResNet50 with augmentation,…
-
Designing an AI Agent for Touhou-Style Bullet Hell Games
Python, PyTorch, Reinforcement Learning, Deep Q-Network (DQN) I made a TouHou simulation game, and recently I add few Algorithms and a simple Q-learning AI to play the game automaticly.…
-
An IoT Test Tool : Development Notes
Python, PyQt6, MQTT, RESTful API, JSON, HTTPS/SSL, MySQL, Docker, Linux, Git Perform 运行表现 I built an IoT testing tool for work. Can’t share it (company rules), but writing …
-
Safety-driven Path Selection Using Reinforcement-Learning in Autonomous Driving
Accepted by RSAE 2025. Abstract—Autonomous driving systems rely heavily on high-quality data in the path selection process to ensure safety and time efficiency during driving.…
-
GBC: Gaussian-Based Colorization and Super-Resolution for 3D Reconstruction
Published by ACM SIGGRAPH VRCAI ’24. In this paper, we introduce GBC, an advanced framework for transforming low-resolution, monochrome video sequences into high-resolution,…
-
BMS^3: Bayesian Modeling Based SwinUNet Segmentation on Self-distillation Architecture
Published by ICONIP 2025. The six-volume set constitutes the refereed proceedings of the 32nd International Conference on Neural Information Processing, ICONIP 2025, held in …











