TRIP: Trainable Region-of-Interest Prediction for Hardware-Efficient Neuromorphic Processing on Event-based Vision

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Arjmand, Cina, Xu, Yingfu, Shidqi, Kevin, Dobrita, Alexandra F., Vadivel, Kanishkan, Detterer, Paul, Sifalakis, Manolis, Yousefzadeh, Amirreza, Tang, Guangzhi
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913404571090944
author Arjmand, Cina
Xu, Yingfu
Shidqi, Kevin
Dobrita, Alexandra F.
Vadivel, Kanishkan
Detterer, Paul
Sifalakis, Manolis
Yousefzadeh, Amirreza
Tang, Guangzhi
author_facet Arjmand, Cina
Xu, Yingfu
Shidqi, Kevin
Dobrita, Alexandra F.
Vadivel, Kanishkan
Detterer, Paul
Sifalakis, Manolis
Yousefzadeh, Amirreza
Tang, Guangzhi
contents Neuromorphic processors are well-suited for efficiently handling sparse events from event-based cameras. However, they face significant challenges in the growth of computing demand and hardware costs as the input resolution increases. This paper proposes the Trainable Region-of-Interest Prediction (TRIP), the first hardware-efficient hard attention framework for event-based vision processing on a neuromorphic processor. Our TRIP framework actively produces low-resolution Region-of-Interest (ROIs) for efficient and accurate classification. The framework exploits sparse events' inherent low information density to reduce the overhead of ROI prediction. We introduced extensive hardware-aware optimizations for TRIP and implemented the hardware-optimized algorithm on the SENECA neuromorphic processor. We utilized multiple event-based classification datasets for evaluation. Our approach achieves state-of-the-art accuracies in all datasets and produces reasonable ROIs with varying locations and sizes. On the DvsGesture dataset, our solution requires 46x less computation than the state-of-the-art while achieving higher accuracy. Furthermore, TRIP enables more than 2x latency and energy improvements on the SENECA neuromorphic processor compared to the conventional solution.
format Preprint
id arxiv_https___arxiv_org_abs_2406_17483
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TRIP: Trainable Region-of-Interest Prediction for Hardware-Efficient Neuromorphic Processing on Event-based Vision
Arjmand, Cina
Xu, Yingfu
Shidqi, Kevin
Dobrita, Alexandra F.
Vadivel, Kanishkan
Detterer, Paul
Sifalakis, Manolis
Yousefzadeh, Amirreza
Tang, Guangzhi
Computer Vision and Pattern Recognition
Image and Video Processing
Neuromorphic processors are well-suited for efficiently handling sparse events from event-based cameras. However, they face significant challenges in the growth of computing demand and hardware costs as the input resolution increases. This paper proposes the Trainable Region-of-Interest Prediction (TRIP), the first hardware-efficient hard attention framework for event-based vision processing on a neuromorphic processor. Our TRIP framework actively produces low-resolution Region-of-Interest (ROIs) for efficient and accurate classification. The framework exploits sparse events' inherent low information density to reduce the overhead of ROI prediction. We introduced extensive hardware-aware optimizations for TRIP and implemented the hardware-optimized algorithm on the SENECA neuromorphic processor. We utilized multiple event-based classification datasets for evaluation. Our approach achieves state-of-the-art accuracies in all datasets and produces reasonable ROIs with varying locations and sizes. On the DvsGesture dataset, our solution requires 46x less computation than the state-of-the-art while achieving higher accuracy. Furthermore, TRIP enables more than 2x latency and energy improvements on the SENECA neuromorphic processor compared to the conventional solution.
title TRIP: Trainable Region-of-Interest Prediction for Hardware-Efficient Neuromorphic Processing on Event-based Vision
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2406.17483