metadata
title: Computervisionobjectdetection
emoji: ⚡
colorFrom: purple
colorTo: pink
sdk: gradio
sdk_version: 5.31.0
app_file: app.py
pinned: false
📷✨ Object Detection Demo
🚀 Overview
Experience real-time, Transformer-powered object detection entirely on CPU.
Upload any image and instantly see bounding boxes, labels, and confidence scores—all wrapped in a sleek Gradio interface.
Core technologies:
• DETR (DEtection TRansformer) for end-to-end CV pipelines
• Hugging Face Transformers for model orchestration
• Gradio Blocks for interactive web UI
• Pillow (PIL) for image annotation
✨ Key Features
| 🔑 Feature | 🔍 Description |
|---|---|
| ⚡ Transformer CV | Uses DETR + ResNet-50 backbone for state-of-the-art accuracy |
| 🕒 Real-Time Inference | Sub-second CPU performance on typical images |
| 🎨 Annotated Output | Red boxes + text overlays for clear visual feedback |
| 📊 Detection Table | Interactive DataFrame of labels & confidence scores |
| ☁️ Cloud-Native Deploy | One-click deploy on free Hugging Face Spaces |
| 🔧 Modular Architecture | Swap models or add filters with minimal code changes |
🏗️ Architecture & Workflow
- Image Upload
User drops in any JPEG/PNG. - DETR Pipeline
pipeline("object-detection", model="facebook/detr-resnet-50") - Post-processing
Draws bounding boxes + labels via PIL. - UI Rendering
Gradio displays the annotated image and a label/score table.
🛠️ Quick Start (Local)
git clone https://github.com/your-username/object-detection-demo.git
cd object-detection-demo
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
python app.py
## Latest Update
- Upgraded DETR model for better accuracy. - May 29, 2025 📝
- Optimized CPU inference for faster detection. 📷 - June 01, 2025 📝
- Enhanced Gradio UI for image uploads. 🎨 - May 31, 2025 📝
- Improved bounding box visualization. - May 30, 2025 📝
**Website**: https://ghostainews.com/
**Discord**: https://discord.gg/BfA23aYz