⚡ 本页包含 AI 生成的分析内容，仅供参考
提出了一款名为DL-VOPU的领域特定深度学习视觉目标处理单元，专门针对视觉目标检测与跟踪（VODT）任务进行架构优化，通过多尺度语义支持实现高能效，适用于移动端智能应用。
years, deep learning-based visual object detection/tracking (VODT) has been widely used in intelligent applications such as autonomous driving, UAV, smart robot and VR/AR. As general AI hardware platforms, GPUs and general AI processors are often used for accelerating VODT. However, without a domain-specific architecture, it is difficult for these processors to achieve high energy efficiency, making them unsuitable for mobile VODT applications. Recently, some dedicated VODT processors have been proposed with improved energy efficiency [1][2][3]. As shown in Fig. 22.7.1, these designs have several issues: 1) they only support a single task (either detection or tracking), 2) they lack full support for multi-scale semantic feature extraction (MSFE)based state-of-the-art VODT frameworks [4], and 3) they do not sufficiently exploit
Yuchuan Gong, Teng Zhang, Hongtao Guo, Xiyuan Liu, Jingxiao Zheng,
Hongqiang Wu, Conghan Jia, Luying Que, Liang Zhou, Liang Chang, Jun Zhou University of Electronic Science and Technology of China, Chengdu, China In the recent