A Multimedia Semantic Analysis SoC (SASoC) with Machine-Learning Engine

📋 论文概要

本文提出了一款用于多媒体语义分析的SoC，集成了机器学习引擎，能够实时处理人脸检测、表情识别、场景分析和物体识别等任务，解决了现有处理器仅关注特征提取而缺乏专门硬件支持的问题。

💡 主要创新点

• 1. 集成机器学习引擎的SoC架构，支持多种多媒体语义分析任务。
• 2. 专用硬件设计，实现实时处理能力，超越仅特征提取的处理器。

🏷 关键词

📄 原文摘要

versatile multimedia applications with semantic processing abilities. Realtime applications, such as face detection, facial-expression recognition, scene analysis [2] and object recognition [3], have become indispensable functionality for Consumer Electronic (CE) products. To deal with complicated video-processing algorithms for multimedia content analysis, many powerful processors have been reported [2-5]. Although these processors can speed up video-processing tasks with massively parallel processing elements, they only focus on the feature-extraction parts, and there is no specialized hardware to support different kinds of advanced machine-learning algorithms, which require extensive computations. In this paper, a Semantic Analysis SoC (SASoC) that accelerates

👥 作者与机构

Tse-Wei Chen, Yi-Ling Chen, Teng-Yuan Cheng, Chi-Sun Tang,

Pei-Kuei Tsung, Tzu-Der Chuang, Liang-Gee Chen, Shao-Yi Chien National Taiwan University, Taipei, Taiwan Advances in semiconductors and developments in machine learning [1] have led to