https://github.com/srp4214/playpulse_player_tracking
https://github.com/srp4214/playpulse_player_tracking
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/srp4214/playpulse_player_tracking
- Owner: SRP4214
- Created: 2025-02-01T07:11:02.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2025-02-01T07:16:04.000Z (5 months ago)
- Last Synced: 2025-02-01T08:21:43.211Z (5 months ago)
- Language: Python
- Size: 13.7 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 🎯 PlayPulse: Analyze like a pro, strategize like a champion!🏆
## 🚀 Overview
**PlayPulse** is an advanced tool that extracts key insights from a recorded gameplay video(ex. Valorant).It leverages YOLOv8, Computer Vision, OCR, and Deep Learning to provide in-depth player performance analysis. The tool presents intuitive visuals, round-wise summaries, and even generates highlight videos! 🎥🔥
## 📥 Input
- Recorded gameplay video 🎮
## 🛠️ Methodology
1. **UI Creation:** Built using Streamlit for an interactive and aesthetic interface.
2. **Frame Extraction:** Extract frames at 2-second intervals from the video.
3. **Object Detection & OCR:** Apply YOLOv8, EasyOCR, and Image Processing to extract event details.
4. **Mapping Events:** Generate a map of [Killer -> Victim -> Timestep] from extracted data.
5. **Player Performance Metrics:** Calculate:
- ✅ Total Kills
- ❌ Total Deaths
- 🎯 Headshots & Headshot %
- ⚖ Kill-Death Ratio
- 🎯 Clutch
6. **Summary:** Generate a timeline of key events 📜
7. **Data Visualization:** Plot Kills vs. Timestep graph 📊
8. **Highlight Video:** Create a short highlight video with key moments 🔥🎬
9. **Weapon Identification:** Use a custom CNN model to classify player’s weapon 🔫
10. **SWOT Analysis:** Generate strengths, weaknesses, opportunities, and threats of the player 📌
11. **AI-Powered Insights:** Use Gemini API to generate engaging content ✨
## 🎨 Features
- 📊 Comprehensive Player Stats
- 📹 Auto-Generated Highlight Video
- 🗺 Chronological Summary
- 🔫 Weapon Detection with CNN
- 🏆 Headshot Accuracy & Performance Trends
- 🤖 AI-Powered Insights & Content Generation
## 🏗️ Tech Stack
- Python 🐍
- Streamlit 🖥️
- YOLOv8 🧠
- EasyOCR 🔍
- OpenCV 👀
- Deep Learning (CNN) 🏗️
- Gemini API
- ffmpeg 📸
- PyTorch 👀
- Makesense AI annotator 📸
## 📸 Screenshots
# Home Page
# Player stats
# Graph
# SWOT
# Highlight
## 🔧 Installation & Usage
1️⃣ Clone the repo:
```bash
git clone https://github.com/SRP4214/PlayPulse_Player_Tracking.git
cd PlayPulse_Player_Tracking
```
2️⃣ Install dependencies:
```bash
pip install -r requirements.txt
```
3️⃣ Run the Streamlit app:
```bash
streamlit run app.py
```
Future Scopes 🚀
1️⃣ Weapon Performance Metrics
Description: Accurate detection for weapons along with its performance and accuracy.
2️⃣ Multi-Game Support
Description: Extend analysis to other games.
3️⃣ Roundwise game analysis
Description: Highlight round wins