ABSTRACT:
This chapter examines the role of Edge AI in enabling real-time multimodal quality assessment by processing AI workloads close to the point where data are generated. It explains how cameras, IoT sensors, environmental information, and product metadata can be processed locally to support immediate quality decisions while reducing latency, bandwidth consumption, cloud dependency, and privacy risks. The chapter discusses Edge AI inferencing, factory machine vision, edge-cloud collaboration, applications in agriculture, food processing, manufacturing, healthcare, pharmaceuticals, and environmental monitoring, and challenges involving computational resources, energy consumption, synchronization, security, reliability, and model maintenance. Future directions include lightweight Vision Transformers, multimodal foundation models, federated learning, neuromorphic computing, energy-efficient accelerators, and adaptive model compression.
Cite this article:
Narendra Kumar Dewangan. Edge AI for Real-Time Multimodal Quality Assessment. Multimodal Artificial Intelligence for Intelligent Quality Assessment. 2026; 1(1): 32-45. DOI: 10.52711/book.anv.9788199786417-05DOI: https://doi.org/10.52711/book.anv.9788199786417-05