Requirements completeness is crucial yet challenging in software engineering, significantly impacting project success. This paper systematically reviews existing research from 2000 to 2024 to consolidate definitions, detection methods, completeness metrics, improvement frameworks, and supporting tools, emphasizing emerging AI-based solutions. Our synthesis identifies the multi-dimensional nature of completeness and reveals opportunities to integrate AI and automation more deeply into the requirements engineering process. We highlight gaps and propose future directions to achieve systematic and scalable completeness checking, emphasizing the transformative potential of Large Language Models (LLMs) in enhancing requirements quality.