# Infinite Possibility > Keep going ## Posts - [Activating Windows 11 Pro on a New PC Using a University KMS License](https://gclinzy.com/activate-windows-11-pro-university-kms/) - [Milestone 3: Training a Lightweight Clarity Model + Exporting to RKNN (RV1126)](https://gclinzy.com/milestone-3-training-a-lightweight-clarity-model-exporting-to-rknn-rv1126/): Milestone: 3 — Train → Export ONNX → Convert RKNN → Ready for On-board Validation 1) What Milestone 3 Is About Milestone 3’s goal is straightforward: At the end of this milestone, I should have a clean, reproducible conversion pipeline and a ready-to-test model on board. 2) Current Status (What’s Done) Training completed (baseline clarity_v0) Evaluation completed (val/test accuracy around ~0.46–0.47; confusion matrix indicates some “collapse”/class bias to middle bins) ONNX export succeeded (confirmed opset = 12) RKNN conversion succeeded (generated clarity_fp.rknn) Next: board-side deployment & runtime verification (SSH connection is unstable on campus network, may temporarily fall back to […] - [Milestone 2 Done: MPIIGaze Subset + Clarity Labeling Pipeline (5-Level Ordinal → Normalized Score)](https://gclinzy.com/milestone-2-mpiigaze-clarity-labeling-pipeline/): What I completed in Milestone 2 This milestone focuses on building a reproducible dataset + labeling pipeline for my capstone project: eye image clarity assessment on RV1126 (RKNN runtime). The goal is not “training the best model yet”, but making sure the entire data pipeline is solid, auditable, and ready for deployment-oriented iteration. Dataset choice: MPIIGaze (subset) I chose MPIIGaze because it provides large-scale, in-the-wild eye-region images with diverse illumination and head poses. To keep iteration fast and the repo lightweight, I sampled a subset: Raw data stays local (not committed to Git). Labeling strategy: 5-level ordinal clarity + normalized […] - [Capstone Kickoff: Connecting Windows, Ubuntu, and RV1126 + Verifying the RKNN Runtime](https://gclinzy.com/capstone-kickoff-rv1126-rknn-runtime/): TL;DR I’m building an eye-image clarity assessment system and deploying it on Rockchip RV1126 (NPU-enabled). This kickoff milestone focuses on engineering fundamentals: multi-host connectivity, board access, and RKNN runtime verification on the target device. Result: the RV1126 can run RKNN inference successfully with stable performance (~28–31 ms avg, ~32–36 FPS in the provided demo). 1. Project Goal and Engineering Roadmap The core goal of this capstone is eye image clarity classification/scoring on an embedded NPU platform, with a fair comparison against traditional focus/clarity metrics. Planned pipeline: The key principle: decouple inference core from UI via a stable API (e.g., JSON/file/socket), […] - [Real-Time Face Gate with OpenCV + OpenVINO ( USB Camera, Rpi4)](https://gclinzy.com/real-time-face-gate-with-opencv-openvino-usb-camera-windows/): This post documents a reproducible mini-project: a real-time “face gate” prototype built with OpenCV for camera capture and OpenVINO for inference. Source code:<CashStolen/rpi-myriad-face-reg> 1)pipeline overview The core pipeline is: This design keeps the code modular and makes it easy to swap models or tune thresholds. 2)environment Install dependencies:(the requirement.txt lies in my git project) 3)models This project uses two OpenVINO IR models (.xml + .bin): The expected path after your download: 4)quick start First of all, get the index of your camera Then, after finishing the necessary connection and settings of your Rpi4 or 5 and camera device, it’s time […] - [2048 Core Algorithm Explained: Compress, Merge, Repeat](https://gclinzy.com/2048-core-algorithm-explained-compress-merge-repeat/): In 2048, the board is simply a fixed 4×4 grid. A clean mental model is a 2D array of integers: 0 means “empty,” and any other value (2, 4, 8, …) is a tile. The UI is just presentation—the real game is the move logic. 1) Forget directions first: reduce everything to “move left” You can understand the entire game by solving one problem: how a single line moves to the left. Every direction can be transformed into that same operation: So the core is a one-line move; directions are just different ways to read/write lines. 2) A one-line move […] - [A Mini Clock-Wallpaper For Programmers With Light Design](https://gclinzy.com/a-mini-clock-wallpaper-for-programmers-with-light-design/): Overview As a person with special mania for dynamic wallpapers, I almost change my wallpaper everyday with wallpaper engine. But due to my running out of my disk space with hundreds of videos or files, I have to find a way out. Then I make a light and mini clock wallpaper with html just like the feature image of this article. If you want to get access of this wallpaper, just two steps are needed to be followed: Then, all you need to do is to click the button “add a wallpaper”, and choose the .html file to add it […] - [How to Batch Convert JPG/PNG Images to WebP Using Python (Command-Line Tool Included)](https://gclinzy.com/how-to-batch-convert-jpg-png-images-to-webp-using-python-command-line-tool-included/): 1. Introduction — Why Convert to WebP? I recently optimized my website’s mobile performance and found that images were the main bottleneck. The PNGs and JPGs are putting great pressure on the mobile processing of mobile users which may cause laggy in your website browsing. But as there are so many images to process, it’s impossible to convert them one by one. So I wrote a Python command-line tool to batch convert JPG/PNG images into WebP while keeping the original directory structure. And don’t worry, it’s just a very simple python script.And of course I’ll try my best to get […] - [第二章：入局](https://gclinzy.com/%e7%ac%ac%e4%ba%8c%e7%ab%a0%ef%bc%9a%e5%85%a5%e5%b1%80/): This content is password-protected. To view it, please enter the password below. Password: - [第一章：山雨欲来](https://gclinzy.com/%e7%ac%ac%e4%b8%80%e7%ab%a0%ef%bc%9a%e5%b1%b1%e9%9b%a8%e6%ac%b2%e6%9d%a5/): This content is password-protected. To view it, please enter the password below. Password: - [引子](https://gclinzy.com/%e5%bc%95%e5%ad%90/): This content is password-protected. To view it, please enter the password below. Password: - [Introduction to Markdown](https://gclinzy.com/introduction-to-markdown/): Hello everyone, in this article, I will introduce the basic usage of Markdown, including editor selection, editting, and a summary of keyboard shortcuts. ATTENTION, the .md teaching text is below the article, which is availble for downloading(no virus indeed (^‿^)) The Green Texts are Links！！！ Editors to choose First of all, you have to choose an editor to support your markdown editting. I myself use Typora so far, which is a powful tool for markdown editting. And you can visit the official website to buy typora or you can try to download the free resources of older versions(u know, I […] ## Pages - [About](https://gclinzy.com/about/): Who aM I? - [Contact](https://gclinzy.com/contact/): CONTACT ME Talk To Me - [Home](https://gclinzy.com/): Infinite Possibility - [Contends](https://gclinzy.com/services/): ConTENDS - [Privacy Policy](https://gclinzy.com/privacy-policy/): Last updated: Nov 15th 2025 Your privacy is important to us. This website is committed to protecting your personal information and ensuring a safe and transparent browsing experience. This Privacy Policy explains what information we collect, how we use it, and how we protect it. 1. Information We Collect 1.1 Information You Provide We may collect personal information that you voluntarily provide through the contact form on our site, including: 1.2 Automatically Collected Information When you visit our site, we may automatically collect non-personal information such as: This information helps us improve website performance and user experience. 2. Cookies and […] ## Optional - [Agent (MCP protocol)](websites-agents.hostinger.com/gclinzy.com/mcp) [comment]: # (Generated by Hostinger Tools Plugin)