Conference paper
Energy-Efficient Application-Specific Instruction-Set Processor for Feature Extraction in Smart Vision Systems
Smart vision sensor systems enable many computer vision applications such as autonomous drones and wearable devices. These battery-powered gadgets have very stringent power consumption requirements. Close-to-sensor feature extraction compressing the full image into descriptive keypoints, is crucial as it allows for several design optimizations.
First, the amount of necessary on-chip memory can be lessened. Second, the volume of data that needs to be exchanged between nodes in Internet of Things (IoT) applications can also be reduced. This work explores the usage of an Application Specific Instruction Set Processor (ASIP) tailored to perform energy-efficient feature extraction in real-time.
The ASIP features a Very Long Instruction Word (VLIW) central core comprising one RV32I RISCV and three vector slots. The on-chip memory sub-system implements parallel multi-bank memories with near-memory data shuffling to enable single-cycle multi-pattern vector access. As a case study, Oriented FAST and Rotated BRIEF (ORB) is used to evaluate the proposed architecture.
We show that the architecture supports VGA-resolution images at 140 Frames-Per-Second (FPS), for one scale, reducing the number of memory accesses by 2 orders of magnitude comparing to other embedded general-purpose architectures.
| Language: | English |
|---|---|
| Publisher: | IEEE |
| Year: | 2021 |
| Pages: | 324-328 |
| Proceedings: | 55<sup>th</sup> Asilomar Conference on Signals, Systems, and Computers |
| ISBN: | 1665458275 , 1665458283 , 9781665458276 , 9781665458283 , 1665458291 and 9781665458290 |
| ISSN: | 10586393 and 25762303 |
| Types: | Conference paper |
| DOI: | 10.1109/IEEECONF53345.2021.9723114 |
| ORCIDs: | Karlsson, Sven |
Energy efficiency FPS Feature extraction Internet of Things applications IoT applications RV32I RISCV Real-time systems Simultaneous localization and mapping System-on-chip VLIW central core Vision sensors Wearable computers application specific instruction set processor battery-powered gadgets close-to-sensor feature extraction computer vision applications energy conservation energy-efficient application-specific instruction-set processor energy-efficient feature extraction feature extraction frames-per-second image resolution image sensors intelligent sensors multibank memories near-memory data on-chip memory oriented FAST and rotated BRIEF power consumption requirements smart vision sensor systems very long instruction word central core vision-based Simultaneous Localization And Mapping (SLAM)
