SMamba: Sparse Mamba for Event-based Object Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Nan, Wang, Yang, Liu, Zhanwen, Li, Meng, An, Yisheng, Zhao, Xiangmo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912197178818560
author Yang, Nan
Wang, Yang
Liu, Zhanwen
Li, Meng
An, Yisheng
Zhao, Xiangmo
author_facet Yang, Nan
Wang, Yang
Liu, Zhanwen
Li, Meng
An, Yisheng
Zhao, Xiangmo
contents Transformer-based methods have achieved remarkable performance in event-based object detection, owing to the global modeling ability. However, they neglect the influence of non-event and noisy regions and process them uniformly, leading to high computational overhead. To mitigate computation cost, some researchers propose window attention based sparsification strategies to discard unimportant regions, which sacrifices the global modeling ability and results in suboptimal performance. To achieve better trade-off between accuracy and efficiency, we propose Sparse Mamba (SMamba), which performs adaptive sparsification to reduce computational effort while maintaining global modeling capability. Specifically, a Spatio-Temporal Continuity Assessment module is proposed to measure the information content of tokens and discard uninformative ones by leveraging the spatiotemporal distribution differences between activity and noise events. Based on the assessment results, an Information-Prioritized Local Scan strategy is designed to shorten the scan distance between high-information tokens, facilitating interactions among them in the spatial dimension. Furthermore, to extend the global interaction from 2D space to 3D representations, a Global Channel Interaction module is proposed to aggregate channel information from a global spatial perspective. Results on three datasets (Gen1, 1Mpx, and eTram) demonstrate that our model outperforms other methods in both performance and efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2501_11971
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SMamba: Sparse Mamba for Event-based Object Detection
Yang, Nan
Wang, Yang
Liu, Zhanwen
Li, Meng
An, Yisheng
Zhao, Xiangmo
Computer Vision and Pattern Recognition
Transformer-based methods have achieved remarkable performance in event-based object detection, owing to the global modeling ability. However, they neglect the influence of non-event and noisy regions and process them uniformly, leading to high computational overhead. To mitigate computation cost, some researchers propose window attention based sparsification strategies to discard unimportant regions, which sacrifices the global modeling ability and results in suboptimal performance. To achieve better trade-off between accuracy and efficiency, we propose Sparse Mamba (SMamba), which performs adaptive sparsification to reduce computational effort while maintaining global modeling capability. Specifically, a Spatio-Temporal Continuity Assessment module is proposed to measure the information content of tokens and discard uninformative ones by leveraging the spatiotemporal distribution differences between activity and noise events. Based on the assessment results, an Information-Prioritized Local Scan strategy is designed to shorten the scan distance between high-information tokens, facilitating interactions among them in the spatial dimension. Furthermore, to extend the global interaction from 2D space to 3D representations, a Global Channel Interaction module is proposed to aggregate channel information from a global spatial perspective. Results on three datasets (Gen1, 1Mpx, and eTram) demonstrate that our model outperforms other methods in both performance and efficiency.
title SMamba: Sparse Mamba for Event-based Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.11971