This paper presents a novel end-to-end system for pedestrian detection using Dynamic Vision Sensors (DVSs). We target applications where multiple sensors transmit data to a local processing unit, which executes a detection algorithm. Our system is composed of (i) a near-chip event filter that compresses and denoises the event stream from the DVS, and (ii) a Binary Neural Network (BNN) detection module that runs on a low-computation edge computing device (in our case a STM32F4 microcontroller). We present the system architecture and provide an end-to-end implementation for pedestrian detection in an office environment. Our implementation reduces transmission size by up to 99.6% compared to transmitting the raw event stream. Our detector is able to perform a detection every 450 ms, with an overall testing F1 score of 83%. The low bandwidth and energy properties of our system make it ideal for IoT applications.
|Original language||English (US)|
|Title of host publication||Proceedings - 2020 IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2020|
|Publisher||IEEE Computer Society|
|Number of pages||6|
|State||Published - Jul 2020|
|Event||19th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2020 - Limassol, Cyprus|
Duration: Jul 6 2020 → Jul 8 2020
|Name||Proceedings of IEEE Computer Society Annual Symposium on VLSI, ISVLSI|
|Conference||19th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2020|
|Period||7/6/20 → 7/8/20|
Bibliographical notePublisher Copyright:
© 2020 IEEE.
- Binary neural networks
- Dynamic vision sensors
- Pedestrian detection