Pushing the Limits of Asynchronous Graph-based Object Detection with Event Cameras

11/22/2022
by   Daniel Gehrig, et al.
0

State-of-the-art machine-learning methods for event cameras treat events as dense representations and process them with conventional deep neural networks. Thus, they fail to maintain the sparsity and asynchronous nature of event data, thereby imposing significant computation and latency constraints on downstream systems. A recent line of work tackles this issue by modeling events as spatiotemporally evolving graphs that can be efficiently and asynchronously processed using graph neural networks. These works showed impressive computation reductions, yet their accuracy is still limited by the small scale and shallow depth of their network, both of which are required to reduce computation. In this work, we break this glass ceiling by introducing several architecture choices which allow us to scale the depth and complexity of such models while maintaining low computation. On object detection tasks, our smallest model shows up to 3.7 times lower computation, while outperforming state-of-the-art asynchronous methods by 7.4 mAP. Even when scaling to larger model sizes, we are 13 outperforming it by 11.5 mAP. As a result, our method runs 3.7 times faster than a dense graph neural network, taking only 8.4 ms per forward pass. This opens the door to efficient, and accurate object detection in edge-case scenarios.

READ FULL TEXT

page 1

page 3

page 7

research
03/31/2022

AEGNN: Asynchronous Event-based Graph Neural Networks

The best performing learning algorithms devised for event cameras work b...
research
04/26/2023

From Chaos Comes Order: Ordering Event Representations for Object Detection

Today, state-of-the-art deep neural networks that process events first c...
research
03/20/2020

Event-based Asynchronous Sparse Convolutional Networks

Event cameras are bio-inspired sensors that respond to per-pixel brightn...
research
08/28/2023

Graph-based Asynchronous Event Processing for Rapid Object Recognition

Different from traditional video cameras, event cameras capture asynchro...
research
10/05/2022

Transferring dense object detection models to event-based data

Event-based image representations are fundamentally different to traditi...
research
06/01/2021

EV-VGCNN: A Voxel Graph CNN for Event-based Object Classification

Event cameras report sparse intensity changes and hold noticeable advant...
research
12/11/2022

Recurrent Vision Transformers for Object Detection with Event Cameras

We present Recurrent Vision Transformers (RVTs), a novel backbone for ob...

Please sign up or login with your details

Forgot password? Click here to reset