"Understanding High-Level Control Systems" matching MCP connectors:

Background removal, upscaling, face restore, document OCR, table extraction, visual understanding.

GPT-4o: GPT-4o (“o” for “omni”) is our versatile, high-intelligence flagship model. It accepts both.

Turns a phone into a camera+Bluetooth remote so AI assistants can see and control any PC.

This is a Unified MCP Server for full document control: PDF manipulation (merge, split, rotate, insert, delete), OCR/data extraction from PDFs/images, PDFs/images conversion, compression, encryption, and watermark management—all without juggling multiple tools.

Vectorise.Me (vectorizer-mcp)

me.vectorise

Converts raster images (JPG/PNG/BMP/WEBP/AVIF) to SVG on the server, then exports to PDF/EPS/DXF/PNG/JPG/WEBP/AVIF. Two tracing engines (direct, planar) plus explicit detail-level, shape-style, and colour controls. Tools: vectorize_image, vectorize_batch, export_svg, get_usage, get_capabilities, get_status. Pay-per-conversion pricing, no subscription

Use OpenMayhem text, embedding and media models with explicit cost controls.

OCR and document understanding: extract text from images, then summarize or translate it.

mcpOAuth

ai.primateintelligence

Video scene understanding for AI agents via the Primate Vision API.