Cloud Vision API

Google Cloud's pre-trained ML models for extracting text, detecting faces, and identifying millions of objects via REST API.

Cloud Vision API integrates Google's computer vision expertise into applications via REST and RPC interfaces. It classifies images into thousands of categories (e.g., "sailboat" or "Eiffel Tower"), detects individual objects, and reads printed or handwritten text via OCR. Developers use it to moderate content by filtering adult or violent imagery and to build searchable image catalogs. The service processes data at scale: it handles millions of requests with sub-second latency while identifying landmarks, logos, and facial expressions (like joy or sorrow) without requiring custom model training.

https://cloud.google.com/vision