metadata

title: Store Entry People Counter
emoji: 🏪
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.23.0
app_file: app.py
pinned: false
license: apache-2.0

Store Entry People Counter

This Hugging Face Space provides an interface to count the number of people entering through a line in a video (e.g., store entrance). It uses YOLOv8 for person detection and tracking.

Features

Real-time person detection using YOLOv8
Object tracking to count unique entries
Visual feedback:
- Green line: Counting threshold
- Blue boxes: Detected people
- Yellow dots: Tracking points
- Red circles: Entry indicators
- Live counter with background

How to Use

Open the Space in your browser
Upload a video file showing people entering (supports most video formats)
Wait for processing (time depends on video length)
View the processed video with counting visualization
See the final count in the results box

Understanding the Output

Green Line: This is the counting threshold. When people cross this line from top to bottom, they are counted
Blue Boxes: Shows detected people in each frame
Yellow Dots: Tracks the position of each person
Red Circles: Appears when someone crosses the line and is counted
Counter: Shows the running total of people who have entered

Technical Details

Uses YOLOv8 for accurate person detection
Implements object tracking to maintain person identity across frames
Counts unique crossings to avoid double-counting
Processes video frames sequentially with progress indication

Limitations

Best results with clear, unobstructed view of people
May not work well in extremely crowded scenes
Video quality affects detection accuracy
Processing time depends on video length and resolution

Credits

Built using: