https://github.com/scriptburn/person-matcher
AI-powered tool to detect and match people across multiple tourist photos using object detection, person re-identification, and similarity scoring. Built with Streamlit, TorchReID, and YOLOv8.
https://github.com/scriptburn/person-matcher
Last synced: 4 months ago
JSON representation
AI-powered tool to detect and match people across multiple tourist photos using object detection, person re-identification, and similarity scoring. Built with Streamlit, TorchReID, and YOLOv8.
- Host: GitHub
- URL: https://github.com/scriptburn/person-matcher
- Owner: scriptburn
- License: other
- Created: 2025-04-10T07:48:53.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-04-10T08:01:06.000Z (10 months ago)
- Last Synced: 2025-06-08T06:10:12.233Z (8 months ago)
- Language: Python
- Size: 9.77 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: license.txt
Awesome Lists containing this project
README
# 🧍♂️ Person Matcher AI
A Streamlit-based AI tool that detects and matches people across multiple images taken from different angles or devices — ideal for tourist photos at places like the Taj Mahal.
[](LICENSE)
---
## ✨ Features
- 📤 Upload multiple images
- 🧠 Detect people using YOLOv8
- 🧬 Extract features via TorchReID
- 📈 Visual match grouping based on cosine similarity
- 🧮 Auto or manual thresholding
- ✅ Label match results as correct/incorrect
- 📦 Save labeled pairs for training dataset
- 📁 Export dataset for COCO/CSV training
- 🐳 Dockerized with Makefile for easy dev/prod workflows
---
## 🖼️ Use Case
> Given Image A and Image B taken at the same place by two different people, the app identifies if both images contain the same individual — even from different angles or cameras.
---
## 🚀 Run Locally (Dev Mode)
```bash
make build
make dev
```
Then visit: [http://localhost:8501](http://localhost:8501)
---
## 🏗️ Project Structure
```
person-matcher/
├── app/
│ ├── detector.py # Person detection (YOLOv8 or simulated)
│ ├── reid.py # Feature extraction and re-ID
├── web/
│ └── app.py # Streamlit frontend
├── data/
│ ├── crops/ # Saved validated crop pairs
│ └── labels/ # labels.jsonl with group metadata
├── uploads/ # Temp uploaded images
├── runtime_crops/ # Temp detection results
├── Dockerfile
├── Makefile
├── requirements.txt
└── .gitignore
```
---
## 📦 Dataset Workflow
1. Upload multiple images
2. Confirm visually matched people
3. Save labeled data
4. Export dataset for training
---
## 📤 Exporting
- Export labeled dataset to COCO or CSV for training (coming soon)
- Future: `make export-coco` or `make export-csv`
---
## 🧪 Tech Stack
- Streamlit
- YOLOv8 (Ultralytics)
- TorchReID
- NumPy + Scikit-learn
- Docker + Makefile
---
## 📜 License
This project is licensed for **non-commercial use only**.
If you wish to use this software in a commercial product or service, please contact the author
---
## 👩💻 Author
Created by Rajneesh ojha— a passionate AI + Web developer from India 🇮🇳
Drop a ⭐ if this project helped you!
---
## 📄 License File
See the [LICENSE](LICENSE) file in this repository for full details.