Tag
This survey paper systematically reviews multimodal code intelligence systems that generate and reason with code from visual inputs like screenshots and charts, categorizing approaches across GUI, scientific visualization, structured graphics, and emerging frameworks while proposing verification-centered future research directions.