DL-VOPU: An Energy-Efficient Domain-Specific Deep-Learning-Based Visual Object Processing Unit Supporting Multi-Scale Semantic Feature Extraction for Mobile Object Detection/Tracking Applications.
Yuchuan GongTeng ZhangHongtao GuoXiyuan LiuJingxiao ZhengHongqiang WuConghan JiaLuying QueLiang ZhouLiang ChangJun ZhouPublished in: ISSCC (2023)
Keyphrases
- deep learning
- domain specific
- processing units
- object detection
- feature extraction
- multiscale
- unsupervised learning
- weakly supervised
- general purpose
- image processing
- object class
- parallel processing
- machine learning
- parallel computing
- computing systems
- real time
- particle filter
- computer vision
- feature vectors
- object categories
- image classification
- relation extraction
- object recognition
- wireless sensor networks
- object tracking
- sensor networks
- appearance model
- mental models
- face recognition
- context aware
- image segmentation
- multi class
- partial occlusion
- semantic information
- semantic features