What frameworks does this skill support?

It supports major vision frameworks including PyTorch, torchvision, Ultralytics (YOLO), Detectron2, MMDetection, and segment-anything (SAM).

Can I use this for edge deployment?

Yes, it includes specific workflows for optimizing models for edge hardware using TensorRT, OpenVINO, CoreML, and TFLite.

How does it handle model optimization?

It guides you through benchmarking baseline performance, exporting to ONNX, and applying quantization techniques (FP16/INT8) to significantly increase inference speed.

Does it help with data labeling?

While it doesn't label images itself, it provides scripts and workflows for converting between annotation formats like COCO, YOLO, and Pascal VOC to prepare datasets for training.

Which detection architectures are included?

The skill covers a wide range of architectures, from real-time models like YOLOv5-v11 and RT-DETR to high-accuracy frameworks like Faster R-CNN and DINO.

Senior Computer Vision Engineer

Name: Senior Computer Vision Engineer
Author: zhangzhang-111-i

byzhangzhang-111-i

0•

数据科学与机器学习

Builds, optimizes, and deploys production-grade computer vision systems for object detection and image segmentation.

This skill provides expert guidance for the entire computer vision lifecycle, from initial dataset preparation and architecture selection to model training and high-performance deployment. It supports industry-standard frameworks like PyTorch and Ultralytics, offering specialized workflows for object detection (YOLO, DETR), instance segmentation (SAM, Mask R-CNN), and model optimization using ONNX and TensorRT. Whether you are building real-time edge applications or high-accuracy cloud services, this skill streamlines the process of creating robust, scalable visual AI systems.

主要功能

01Comparative analysis of CNN vs. Vision Transformer architectures

02End-to-end object detection and segmentation pipeline generation

03Real-time inference benchmarking and performance tuning

04Advanced model optimization for ONNX, TensorRT, and OpenVINO

05Automated dataset preparation and augmentation workflows

060 GitHub stars

使用场景

01Building automated image annotation and custom dataset pipelines

02Optimizing heavy deep learning models for high-throughput production serving

03Developing real-time object detection for edge devices or mobile apps

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zhangzhang-111-i/claude-skills111 senior-computer-vision

For use in Claude.ai and ChatGPT

主要功能

01Comparative analysis of CNN vs. Vision Transformer architectures

02End-to-end object detection and segmentation pipeline generation

03Real-time inference benchmarking and performance tuning

04Advanced model optimization for ONNX, TensorRT, and OpenVINO

05Automated dataset preparation and augmentation workflows

060 GitHub stars

使用场景

01Building automated image annotation and custom dataset pipelines

02Optimizing heavy deep learning models for high-throughput production serving

03Developing real-time object detection for edge devices or mobile apps

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zhangzhang-111-i/claude-skills111 senior-computer-vision

For use in Claude.ai and ChatGPT