The field of AI that lets software read images and video, from recognizing objects and faces to inspecting parts on a production line.
Computer Vision enables AI to extract usable information from video feeds, satellite imagery, and scanned documents. Combined with LLMs in a Multi-Modal AI system, it lets enterprises query their physical infrastructure in natural language, for example by asking the system to 'Identify structural anomalies in the pipeline footage'.
See also Multi-Modal AI, Large Language Model, Fine-Tuning and Model Inference.