KinderSort v1.2

KinderSort v1.2 — Documentation & Accuracy Corrections

What's New

Bilingual README Improvements

• Added Step 7 to the screenshot walkthrough — shows the spinning timer display so teachers know what to expect during a long sort
• Added processing time guidance: sorting 500 photos typically takes 8–15 minutes on a standard PC (v1.1's higher-accuracy CNN detection is worth the wait)
• Corrected face matching threshold shown in docs from 0.5 → 0.55 (matches actual v1.1 code)

Chinese README (简体中文) Fixes

• Corrected an outdated note that said photos placed directly in the Events root folder would be skipped — they are in fact supported; the folder name is used as the event name
• Added the same processing time guidance in Chinese

---

No Code Changes in This Release

All face recognition, GUI, and sorting logic is unchanged from v1.1. This release only updates documentation.

---

Download

Download KinderSort.exe from the Assets section below — same binary as v1.1.

v1.0.1

KinderSort v1.0 — First Release

Download: KinderSort.exe (no installation required)

What it does

KinderSort automatically sorts kindergarten event photos into individual student folders using
offline face recognition. No internet connection, no coding knowledge needed — just double-click and
go.

Features

• Automatic face matching — scans event photos and copies each photo into the matching student's
  folder
• Group photo support — if multiple students appear in one photo, it is copied to all of them
• Flexible events folder — works whether photos are inside event subfolders or placed directly in
  the folder
• Unmatched photos — photos with no face, blurry faces, or unknown faces go to _unmatched/
  automatically
• Fully offline — no cloud upload, no internet, no GPU required
• Audit log — every action is recorded in kindersort_log.txt in the output folder
• Safe — photos are always copied, never moved or deleted

---

How to use

1. Download KinderSort.exe below
2. Double-click to open — no installation needed
3. Select your Reference Photos folder (one photo per student, named by student name e.g. Ali.jpg)
4. Select your Events Folder (photos from school events)
5. Select an Output Folder where sorted results will be saved
6. Click Start Sorting

See the full guidebook.md for step-by-step instructions with screenshots.

---

System requirements

• Windows 10 or 11
• No Python, no installation needed
• No GPU is needed

---

Known limitations

• Accuracy depends on photo quality — clear, front-facing reference photos work best
• Small faces in group shots may not be detected
• Processing speed depends on PC — allow up to 15 minutes for 500 photos on older machines

---

▎ Built with face_recognition, dlib, Pillow, tkinter, PyInstaller

---

KinderSort v1.0.0 — First Public Release

KinderSort is an offline Windows app that helps kindergarten teachers sort event photos into student folders using face recognition.

What’s included

• KinderSort.exe (Windows, no Python installation required)

Highlights

• Fully offline processing (privacy-friendly)
• Simple GUI workflow: Reference / Events / Output
• Group photos copied to all matched students
• Unmatched photos saved to _unmatched
• Detailed log output in kindersort_log.txt

Notes

• Recommended for Windows 10/11
• See README for setup tips and troubleshooting