mcp-server-cvdlt

mcp-server-cvdlt

3

This project is a Python server implementing the Model Context Protocol (MCP) for advanced image processing tasks such as object detection, segmentation, and pose estimation. It utilizes well-known models like YOLOv10 and YOLOv8 for these capabilities, offering support for both local and network image inputs and integration with client interactions.

MCP Server for CVDLT (Computer Vision & Deep Learning Tools)

  • Based on Ultralytics and Model Context Protocol Python SDK.
  • Implements image object detection, segmentation, and pose estimation.
  • Supports YOLOv10 for object detection, YOLOv8 for segmentation, and Ultralytics SAM for full image segmentation.
  • Features local and network image inputs, MCP tool integration, stdio, and SSE transport protocols.
  • Planned features include 3D detection, AIGC, Denso estimation, and DL model deployment.

Features

  • Detect objects using YOLOv10.
  • Segment objects using YOLOv8.
  • Segment entire images using Ultralytics SAM.
  • Estimate human poses using YOLOv8.

TODO

  • 3D Detection
  • AIGC (GAN, Diffusion)
  • Denso Estimation
  • Deploy DL Models