What frameworks are integrated into this skill?

The skill is built around the PyTorch ecosystem, including libraries like torchvision, Ultralytics, Detectron2, MMDetection, and MMSegmentation.

Can this skill help with model deployment and optimization?

Yes, it includes specialized workflows for exporting models to ONNX, optimizing with TensorRT, and applying quantization for deployment on NVIDIA GPUs, Intel CPUs, and mobile devices.

Does it handle dataset format conversions?

Absolutely. It assists in converting and validating annotations between COCO, Pascal VOC, YOLO, and LabelMe formats, as well as setting up augmentation pipelines.

Which computer vision architectures are supported?

It supports a wide range of architectures including the YOLO family (v5-v11), Faster R-CNN, DETR, Mask R-CNN, Segment Anything (SAM), and Vision Transformers (ViT).

Senior Computer Vision Engineering

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

byzhangzhang-111-i

0•

데이터 과학 및 ML

Architects and deploys production-grade computer vision systems for object detection, image segmentation, and advanced visual AI applications.

This skill provides end-to-end guidance for building high-performance visual AI systems, covering everything from custom dataset preparation and architecture selection to advanced model optimization for production. It enables engineers to implement state-of-the-art architectures like YOLOv11, DETR, and Segment Anything (SAM) while ensuring efficient deployment via ONNX and TensorRT across cloud and edge devices. Whether you are building real-time detection pipelines or fine-tuning vision transformers, this skill offers the specialized expertise needed to bridge the gap between research and production.

주요 기능

010 GitHub stars

02Automated dataset preparation, format conversion (COCO/VOC), and augmentation

03Model optimization for production using ONNX, TensorRT, and INT8 quantization

04End-to-end object detection pipelines using YOLO, Faster R-CNN, and RT-DETR

05Advanced segmentation workflows including Instance, Semantic, and Panoptic tasks

06Architecture selection guides comparing CNN and Vision Transformer trade-offs

사용 사례

01Optimizing high-latency research models for high-throughput production serving

02Developing custom medical imaging segmentation models using SAM or Mask R-CNN

03Building real-time industrial inspection systems for edge deployment

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

주요 기능

010 GitHub stars

02Automated dataset preparation, format conversion (COCO/VOC), and augmentation

03Model optimization for production using ONNX, TensorRT, and INT8 quantization

04End-to-end object detection pipelines using YOLO, Faster R-CNN, and RT-DETR

05Advanced segmentation workflows including Instance, Semantic, and Panoptic tasks

06Architecture selection guides comparing CNN and Vision Transformer trade-offs

사용 사례

01Optimizing high-latency research models for high-throughput production serving

02Developing custom medical imaging segmentation models using SAM or Mask R-CNN

03Building real-time industrial inspection systems for edge deployment

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