senior-computer-vision

Guide building, training, and deploying computer vision models with PyTorch and ONNX.

Updated Feb 27, 2026
One-click install
npx skills add https://github.com/amanhsn/flyerbuild --skill senior-computer-vision-amanhsn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: senior-computer-vision
Source: https://github.com/amanhsn/flyerbuild/tree/main/.cursor/skills/senior-computer-vision
Command: npx skills add https://github.com/amanhsn/flyerbuild --skill senior-computer-vision-amanhsn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires onnx, tensorrt, openvino, torchvision, ultralytics, detectron2, mmdetection, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the complex process of developing, optimizing, and deploying computer vision models for tasks like object detection and image segmentation.

Core Features & Use Cases

  • End-to-End Pipelines: Guides you through dataset preparation, model training, and production deployment.
  • Architecture Selection: Recommends optimal models (YOLO, ViT, etc.) based on your specific accuracy and speed requirements.
  • Model Optimization: Provides tools and strategies for exporting and accelerating models using ONNX, TensorRT, and other runtimes.
  • Use Case: You need to build a real-time object detection system for a retail environment. This Skill helps you select the best YOLO variant, prepare your custom dataset, train the model, and optimize it for deployment on an edge device.

Quick Start

Use the senior-computer-vision skill to generate a training configuration for a YOLOv8 object detection model using your dataset located in '/path/to/your/dataset'.

Frequently Asked Questions about senior-computer-vision

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy a computer vision model for real-time object detection on an edge device?

TensorRT and ONNX optimize deep learning models by converting them into serialized engines for hardware-specific acceleration. This reduces inference latency and improves throughput for production computer vision deployments.

What's the best way to choose between YOLO and Vision Transformers for image segmentation?

Choosing between YOLO and Vision Transformers for image segmentation depends on your accuracy and speed requirements. YOLO variants are optimal for real-time speed, while Transformers excel in complex feature extraction tasks.

Can I use Detectron2 and MMDetection within the same PyTorch training pipeline?

You can integrate Detectron2 and MMDetection within PyTorch pipelines to manage datasets and train segmentation models. Both frameworks provide modular components for building custom computer vision architectures.

How do I prepare a custom dataset for training an object detection model?

Preparing a custom dataset for object detection involves structuring images and annotations to match framework requirements like Ultralytics. Proper dataset management ensures accurate model training and validation.

Does OpenVINO support model optimization for CNNs and Transformers?

OpenVINO supports model optimization for both CNNs and Transformers by converting them into an intermediate representation. This allows efficient execution across various Intel hardware architectures for computer vision tasks.