skills.rest
AgLyx3
AgLyx3COMMUNITY
Explore all skills by this author →
GitHub

senior-computer-vision

Implement end-to-end computer vision pipelines for object detection and segmentation.

Updated Mar 21, 2026
Search Tags:#pytorch#production-deployment#model-selection#onnx#computer-vision#tensorrt#dataset-preparation
One-click install
npx skills add https://github.com/AgLyx3/My-Note-App-Not-Just-a-Note-App --skill senior-computer-vision-aglyx3
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: senior-computer-vision
Source: https://github.com/AgLyx3/My-Note-App-Not-Just-a-Note-App/tree/main/.cursor/skills/engineering-team/senior-computer-vision
Command: npx skills add https://github.com/AgLyx3/My-Note-App-Not-Just-a-Note-App --skill senior-computer-vision-aglyx3

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires numpy, onnx, onnxruntime, torch, opencv-python, pyyaml, coremltools, and includes scripts (resource) and references (resource) components.

What problem does it solve?

The Senior Computer Vision skill addresses the complexity of designing, training, and deploying production-grade computer vision systems, covering object detection, instance and semantic segmentation, and efficient serving across cloud, edge, and embedded environments.

Core Features & Use Cases

  • End-to-end CV pipeline guidance from architecture selection to deployment across platforms.
  • Comprehensive coverage of CNN and Vision Transformer architectures, optimization with ONNX/TensorRT, and integration with PyTorch ecosystems (Ultralytics, Detectron2, MMDetection).
  • Real-world use cases include building detection/segmentation pipelines, model optimization, dataset preparation, and scalable serving.

Quick Start

Configure a minimal, production-ready CV workflow by selecting architectures, preparing data, and provisioning deployment targets for rapid prototyping.

Frequently Asked Questions about senior-computer-vision

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

FAQPage Schema
How do I build an end-to-end computer vision pipeline for object detection?▼

Building an end-to-end computer vision pipeline involves model selection, dataset preparation, training with PyTorch frameworks like Ultralytics, and optimizing via ONNX or TensorRT for scalable deployment across cloud, edge, and embedded devices.

How do I optimize and deploy PyTorch models for edge devices?▼

Optimizing PyTorch models for edge devices requires exporting trained networks to ONNX format and leveraging TensorRT or OpenVINO. This ensures efficient inference and scalable serving on embedded hardware.

Does this computer vision workflow support Vision Transformer architectures?▼

Yes, the computer vision workflow supports both CNN backbones and Vision Transformer architectures. It integrates with PyTorch ecosystems like Detectron2 and MMDetection to train and deploy these models effectively.

What's the best way to convert a segmentation model to ONNX?▼

Converting a segmentation model to ONNX involves using PyTorch and coremltools to export the trained network. This prepares the model for optimized inference using ONNX Runtime or TensorRT in production environments.

Can I use Ultralytics and Detectron2 for dataset preparation and training?▼

Yes, you can use Ultralytics and Detectron2 for dataset preparation and training. These popular PyTorch frameworks are fully supported to build, train, and validate object detection and segmentation pipelines.

Related Skills

Explore More Software Engineering Skills→
SignmanalSignmanal
1

segment-anything-model

Segment arbitrary objects in images using point, box, or mask prompts.

Community
Intermediate
NVIDIANVIDIA
13.2k

trt-onnx-quickstart

Convert ONNX models to TensorRT engines and verify numerical parity.

Official
Advanced
4lerman4lerman

senior-computer-vision

Design, train, and deploy computer vision models with PyTorch and ONNX.

Community
Advanced
salamientarksalamientark

senior-computer-vision

Design and deploy computer vision models with ONNX, TensorRT, and OpenVINO backends.

Community
Advanced