Vision + Image Hub
Everything Meridian offers for computer vision and image processing. From raw ingestion through generation and editing, to recipe-specific classification, extraction, and human review.
Vision
Computer vision pipeline overview, model architecture, and inference runtime.
Image Input
Ingestion endpoints, accepted formats, size limits, and preprocessing transforms.
Image Generation
Text-to-image and image-to-image generation via Meridian diffusion models.
Image Edits
Inpainting, outpainting, background removal, and mask-based editing APIs.
Image Workflow
Chaining vision steps into reusable pipelines with conditional branching.
Recipe Image Classifier
Fine-tuned classifier that labels recipe photos by dish type, cuisine, and plating style.
Recipe Data Extraction
OCR and structured parsing that pulls ingredients, steps, and cook times from recipe images.
Recipe KYC Review
Human-in-the-loop review queue for flagged recipe submissions and edge-case verification.
Getting started
New to the vision hub? Begin with Vision for an architectural overview, then explore individual modules as your use case demands.