TENet: Targetness Entanglement Incorporating with Multi-Scale Pooling and Mutually-Guided Fusion for RGB-E Object Tracking

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shao, Pengcheng, Xu, Tianyang, Tang, Zhangyong, Li, Linze, Wu, Xiao-Jun, Kittler, Josef
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910438704283648
author Shao, Pengcheng
Xu, Tianyang
Tang, Zhangyong
Li, Linze
Wu, Xiao-Jun
Kittler, Josef
author_facet Shao, Pengcheng
Xu, Tianyang
Tang, Zhangyong
Li, Linze
Wu, Xiao-Jun
Kittler, Josef
contents There is currently strong interest in improving visual object tracking by augmenting the RGB modality with the output of a visual event camera that is particularly informative about the scene motion. However, existing approaches perform event feature extraction for RGB-E tracking using traditional appearance models, which have been optimised for RGB only tracking, without adapting it for the intrinsic characteristics of the event data. To address this problem, we propose an Event backbone (Pooler), designed to obtain a high-quality feature representation that is cognisant of the innate characteristics of the event data, namely its sparsity. In particular, Multi-Scale Pooling is introduced to capture all the motion feature trends within event data through the utilisation of diverse pooling kernel sizes. The association between the derived RGB and event representations is established by an innovative module performing adaptive Mutually Guided Fusion (MGF). Extensive experimental results show that our method significantly outperforms state-of-the-art trackers on two widely used RGB-E tracking datasets, including VisEvent and COESOT, where the precision and success rates on COESOT are improved by 4.9% and 5.2%, respectively. Our code will be available at https://github.com/SSSpc333/TENet.
format Preprint
id arxiv_https___arxiv_org_abs_2405_05004
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TENet: Targetness Entanglement Incorporating with Multi-Scale Pooling and Mutually-Guided Fusion for RGB-E Object Tracking
Shao, Pengcheng
Xu, Tianyang
Tang, Zhangyong
Li, Linze
Wu, Xiao-Jun
Kittler, Josef
Computer Vision and Pattern Recognition
There is currently strong interest in improving visual object tracking by augmenting the RGB modality with the output of a visual event camera that is particularly informative about the scene motion. However, existing approaches perform event feature extraction for RGB-E tracking using traditional appearance models, which have been optimised for RGB only tracking, without adapting it for the intrinsic characteristics of the event data. To address this problem, we propose an Event backbone (Pooler), designed to obtain a high-quality feature representation that is cognisant of the innate characteristics of the event data, namely its sparsity. In particular, Multi-Scale Pooling is introduced to capture all the motion feature trends within event data through the utilisation of diverse pooling kernel sizes. The association between the derived RGB and event representations is established by an innovative module performing adaptive Mutually Guided Fusion (MGF). Extensive experimental results show that our method significantly outperforms state-of-the-art trackers on two widely used RGB-E tracking datasets, including VisEvent and COESOT, where the precision and success rates on COESOT are improved by 4.9% and 5.2%, respectively. Our code will be available at https://github.com/SSSpc333/TENet.
title TENet: Targetness Entanglement Incorporating with Multi-Scale Pooling and Mutually-Guided Fusion for RGB-E Object Tracking
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.05004