# NQvision
NQvision is a powerful library built around Ultralytics models in ONNX format, designed to simplify the development of AI-driven object detection and tracking solutions. It transforms complex computer vision capabilities into an accessible, production-ready solution that revolutionizes how organizations approach real-time monitoring and security.
## 🚀 Features
### Core Capabilities
- **ONNX Model Integration**: Seamless integration with Ultralytics models
- **Real-Time Object Detection**: Optimized for immediate recognition and action
- **Continuous Object Tracking**: Advanced tracking maintaining object identities across frames
- **High-Performance Processing**: Efficient operation on both CPU and GPU
- **Customizable Detection Settings**: Adjustable confidence thresholds and tracking configurations
- **Scalable Architecture**: Handles multiple video feeds simultaneously
### Event Management
- **Real-Time Event Alerts**: Instant notification system for critical detections
- **Event Aggregation**: Intelligent clustering of detections to reduce false positives
- **Customizable Criteria**: Configurable detection thresholds and frequency parameters
- **High-Confidence Alerts**: Aggregated detection within defined time windows
- **Scalable Event Management**: Suitable for both small setups and enterprise deployments
## 💫 Key Benefits
### Unmatched Flexibility
- Universal Ultralytics Compatibility
- Expanding Architecture Support
- Adaptable Integration with existing security infrastructure
### Enterprise-Grade Performance
- Scalable from single-camera setups to city-wide deployments
- Resource-optimized processing
- Built for 24/7 mission-critical environments
### Revolutionary Features
- Intelligent Tracking across camera views
- Event Streaming with customizable detection criteria
- Automated Response System
- Multi-Camera Coordination
- Seamless handling of multiple video streams
## 🎯 Impact
### For Developers
- Eliminates the need to develop intricate AI pipelines from scratch
- Provides a ready-to-use framework for advanced surveillance
- Customizable settings and real-time capabilities
- Implement AI detection without deep AI expertise
### For Companies
- Accelerate deployment of AI-driven surveillance systems
- Minimize development costs
- Improve system reliability
- Handle complex, large-scale environments
- Event-driven architecture for prompt action on high-risk detections
## ⚡ Quick Start
### Dependencies
To install NQvision Dependencies, follow these steps:
- Install NQvision requirements found in ‘requirements.txt’:
```bash
pip install -r requirements.txt
```
- install onnxruntime :
- For cpu only inference :
```bash
pip install onnxruntime
```
- For gpu accelerated inference
```bash
pip install onnxruntime-gpu --extra-index-url https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/onnxruntime-cuda-12/pypi/simple/
For CUDA 11.X (default):
pip install onnxruntime-gpu
```
## Verifying the Installation
To verify that NQvision is installed correctly, run the following Python code:
```python
from NQvision.core import NQvisionCore, ModelConfig
# Create a basic configuration
config = ModelConfig(input_size=(640, 640), confidence_threshold=0.4)
# Initialize NQvisionCore (replace with your model path)
detector = NQvisionCore("path/to/model/model.onnx", config)
print("NQvision initialized successfully!")
```
If you see the success message without any errors, NQvision is installed and ready to use.
## 🔄 Current Support
- Currently supporting models such as rtlder
- Designed for future expansion
- Regular updates and expanding capabilities
## 🛠 Integration
### Deployment Features
- Rapid deployment: Operational in minutes
- Immediate enhancement of surveillance capabilities
- Minimal training requirements
- Intuitive system for security teams
### System Requirements
- Compatible with existing cameras and systems
- Supports both CPU and GPU processing
- Scalable for various deployment sizes
## 🔮 Future Development
NQvision is designed for continuous evolution, with plans to:
- Adopt additional models and architectures
- Expand ecosystem support
- Regular feature updates
- Enhanced capabilities based on community feedback
## 📝 License
[License details to be added]
## 🤝 Contributing
[Contribution guidelines to be added]
## 📞 Support
[\[Support\]](https://www.linkedin.com/company/neuron-q/)
---
Developed by Neuron Q | Making advanced surveillance technology accessible
Raw data
{
"_id": null,
"home_page": "https://neuronq.io",
"name": "NQvision",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": "computer vision, object detection, tracking, CUDA, enterprise",
"author": "Neuron Q",
"author_email": "debbichi.raki@neuronq.io",
"download_url": "https://files.pythonhosted.org/packages/9a/61/03f17911a40496128a99b71f83cd59bbeecf71629259ac9c772be3fc4e83/nqvision-0.1.2.tar.gz",
"platform": null,
"description": "# NQvision\n\nNQvision is a powerful library built around Ultralytics models in ONNX format, designed to simplify the development of AI-driven object detection and tracking solutions. It transforms complex computer vision capabilities into an accessible, production-ready solution that revolutionizes how organizations approach real-time monitoring and security.\n\n## \ud83d\ude80 Features\n\n### Core Capabilities\n\n- **ONNX Model Integration**: Seamless integration with Ultralytics models\n- **Real-Time Object Detection**: Optimized for immediate recognition and action\n- **Continuous Object Tracking**: Advanced tracking maintaining object identities across frames\n- **High-Performance Processing**: Efficient operation on both CPU and GPU\n- **Customizable Detection Settings**: Adjustable confidence thresholds and tracking configurations\n- **Scalable Architecture**: Handles multiple video feeds simultaneously\n\n### Event Management\n\n- **Real-Time Event Alerts**: Instant notification system for critical detections\n- **Event Aggregation**: Intelligent clustering of detections to reduce false positives\n- **Customizable Criteria**: Configurable detection thresholds and frequency parameters\n- **High-Confidence Alerts**: Aggregated detection within defined time windows\n- **Scalable Event Management**: Suitable for both small setups and enterprise deployments\n\n## \ud83d\udcab Key Benefits\n\n### Unmatched Flexibility\n\n- Universal Ultralytics Compatibility\n- Expanding Architecture Support\n- Adaptable Integration with existing security infrastructure\n\n### Enterprise-Grade Performance\n\n- Scalable from single-camera setups to city-wide deployments\n- Resource-optimized processing\n- Built for 24/7 mission-critical environments\n\n### Revolutionary Features\n\n- Intelligent Tracking across camera views\n- Event Streaming with customizable detection criteria\n- Automated Response System\n- Multi-Camera Coordination\n- Seamless handling of multiple video streams\n\n## \ud83c\udfaf Impact\n\n### For Developers\n\n- Eliminates the need to develop intricate AI pipelines from scratch\n- Provides a ready-to-use framework for advanced surveillance\n- Customizable settings and real-time capabilities\n- Implement AI detection without deep AI expertise\n\n### For Companies\n\n- Accelerate deployment of AI-driven surveillance systems\n- Minimize development costs\n- Improve system reliability\n- Handle complex, large-scale environments\n- Event-driven architecture for prompt action on high-risk detections\n\n## \u26a1 Quick Start\n\n### Dependencies\n\nTo install NQvision Dependencies, follow these steps:\n\n- Install NQvision requirements found in \u2018requirements.txt\u2019:\n\n```bash\npip install -r requirements.txt\n```\n\n- install onnxruntime :\n - For cpu only inference :\n ```bash\n pip install onnxruntime\n ```\n - For gpu accelerated inference\n ```bash\n pip install onnxruntime-gpu --extra-index-url https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/onnxruntime-cuda-12/pypi/simple/\n\n For CUDA 11.X (default):\n pip install onnxruntime-gpu\n ```\n\n## Verifying the Installation\n\nTo verify that NQvision is installed correctly, run the following Python code:\n\n```python\nfrom NQvision.core import NQvisionCore, ModelConfig\n\n# Create a basic configuration\nconfig = ModelConfig(input_size=(640, 640), confidence_threshold=0.4)\n\n# Initialize NQvisionCore (replace with your model path)\ndetector = NQvisionCore(\"path/to/model/model.onnx\", config)\n\nprint(\"NQvision initialized successfully!\")\n```\n\nIf you see the success message without any errors, NQvision is installed and ready to use.\n\n## \ud83d\udd04 Current Support\n\n- Currently supporting models such as rtlder\n- Designed for future expansion\n- Regular updates and expanding capabilities\n\n## \ud83d\udee0 Integration\n\n### Deployment Features\n\n- Rapid deployment: Operational in minutes\n- Immediate enhancement of surveillance capabilities\n- Minimal training requirements\n- Intuitive system for security teams\n\n### System Requirements\n\n- Compatible with existing cameras and systems\n- Supports both CPU and GPU processing\n- Scalable for various deployment sizes\n\n## \ud83d\udd2e Future Development\n\nNQvision is designed for continuous evolution, with plans to:\n\n- Adopt additional models and architectures\n- Expand ecosystem support\n- Regular feature updates\n- Enhanced capabilities based on community feedback\n\n## \ud83d\udcdd License\n\n[License details to be added]\n\n## \ud83e\udd1d Contributing\n\n[Contribution guidelines to be added]\n\n## \ud83d\udcde Support\n\n[\\[Support\\]](https://www.linkedin.com/company/neuron-q/)\n\n---\n\nDeveloped by Neuron Q | Making advanced surveillance technology accessible\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "A library to simplify the development of AI-driven object detection and monitoring solutions.",
"version": "0.1.2",
"project_urls": {
"Homepage": "https://neuronq.io"
},
"split_keywords": [
"computer vision",
" object detection",
" tracking",
" cuda",
" enterprise"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "8e0ab7492354652a5293e3b9402d420613172118c2cfe548080a40393afb00bb",
"md5": "5c6dd85a29cba3a0da544690ed6167f1",
"sha256": "f9f43e7f82ddd1ec58f1271525a8a0ab9c4e21acd493518051624f5a0de4764d"
},
"downloads": -1,
"filename": "NQvision-0.1.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "5c6dd85a29cba3a0da544690ed6167f1",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 10106,
"upload_time": "2024-10-28T18:38:24",
"upload_time_iso_8601": "2024-10-28T18:38:24.865719Z",
"url": "https://files.pythonhosted.org/packages/8e/0a/b7492354652a5293e3b9402d420613172118c2cfe548080a40393afb00bb/NQvision-0.1.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "9a6103f17911a40496128a99b71f83cd59bbeecf71629259ac9c772be3fc4e83",
"md5": "5520868388ae659c7133ec99dec766b3",
"sha256": "9cf543c049d0b13f4772d337d1c6b026570b684b88cdf7d2b23668dcfb43930a"
},
"downloads": -1,
"filename": "nqvision-0.1.2.tar.gz",
"has_sig": false,
"md5_digest": "5520868388ae659c7133ec99dec766b3",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 11234,
"upload_time": "2024-10-28T18:38:25",
"upload_time_iso_8601": "2024-10-28T18:38:25.790379Z",
"url": "https://files.pythonhosted.org/packages/9a/61/03f17911a40496128a99b71f83cd59bbeecf71629259ac9c772be3fc4e83/nqvision-0.1.2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-10-28 18:38:25",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "nqvision"
}
JSON