Performance estimation for the memristor-based computing-in-memory implementation of extremely factorized network for real-time and low-power semantic segmentation.
Shuai DongZhen FanYihong ChenKaihui ChenMinghui QinMin ZengXubing LuGuofu ZhouXingsen GaoJun-Ming LiuPublished in: Neural Networks (2023)
Keyphrases
- low power
- vlsi architecture
- low cost
- semantic segmentation
- high speed
- signal processor
- real time
- power consumption
- low power consumption
- cmos technology
- vlsi implementation
- power dissipation
- quantum mechanics
- conditional random fields
- ultra low power
- signal processing
- computer vision
- weakly supervised
- scene classification
- mixed signal
- label transfer
- superpixels
- object class
- image representation