Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space Targets

Estimating the 6D pose of a space target is an intricate task due to factors such as occlusions, changes in visual appearance, and background clutter. Accurate pose determination requires robust algorithms capable of handling these complexities while maintaining reliability under various environment...

Full description

Saved in:
Bibliographic Details
Main Authors: Xu Gu, Xi Yang, Hong Liu, Dong Yang
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/16/22/4138
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846152560261464064
author Xu Gu
Xi Yang
Hong Liu
Dong Yang
author_facet Xu Gu
Xi Yang
Hong Liu
Dong Yang
author_sort Xu Gu
collection DOAJ
description Estimating the 6D pose of a space target is an intricate task due to factors such as occlusions, changes in visual appearance, and background clutter. Accurate pose determination requires robust algorithms capable of handling these complexities while maintaining reliability under various environmental conditions. Conventional pose estimation for space targets unfolds in two stages: establishing 2D–3D correspondences using keypoint detection networks and 3D models, followed by pose estimation via the perspective-n-point algorithm. The accuracy of this process hinges critically on the initial keypoint detection, which is currently limited by predominantly singular-scale detection techniques and fails to exploit sufficient information. To tackle the aforementioned challenges, we propose an adaptive dual-stream aggregation network (ADSAN), which enables the learning of finer local representations and the acquisition of abundant spatial and semantic information by merging features from both inter-layer and intra-layer perspectives through a multi-grained approach, consolidating features within individual layers and amplifying the interaction of distinct resolution features between layers. Furthermore, our ADSAN implements the selective keypoint focus module (SKFM) algorithm to alleviate problems caused by partial occlusions and viewpoint alterations. This mechanism places greater emphasis on the most challenging keypoints, ensuring the network prioritizes and optimizes its learning around these critical points. Benefiting from the finer and more robust information of space objects extracted by the ADSAN and SKFM, our method surpasses the SOTA method PoET (5.8°, 8.1°/0.0351%, 0.0744%) by 0.5°, 0.9°, and 0.0084%, 0.0354%, achieving 5.3°, 7.2° in rotation angle errors and 0.0267%, 0.0390% in normalized translation errors on the Speed and SwissCube datasets, respectively.
format Article
id doaj-art-43d289f18d3e41409425b37be2946e6b
institution Kabale University
issn 2072-4292
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-43d289f18d3e41409425b37be2946e6b2024-11-26T18:19:40ZengMDPI AGRemote Sensing2072-42922024-11-011622413810.3390/rs16224138Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space TargetsXu Gu0Xi Yang1Hong Liu2Dong Yang3State Key Laboratory of Integrated Services Networks, School of Telecommunications Engineering, Xidian University, Xi’an 710071, ChinaState Key Laboratory of Integrated Services Networks, School of Telecommunications Engineering, Xidian University, Xi’an 710071, ChinaSpace Engineering University, Beijing 101416, ChinaXi’an Institute of Space Radio Technology, Xi’an 710100, ChinaEstimating the 6D pose of a space target is an intricate task due to factors such as occlusions, changes in visual appearance, and background clutter. Accurate pose determination requires robust algorithms capable of handling these complexities while maintaining reliability under various environmental conditions. Conventional pose estimation for space targets unfolds in two stages: establishing 2D–3D correspondences using keypoint detection networks and 3D models, followed by pose estimation via the perspective-n-point algorithm. The accuracy of this process hinges critically on the initial keypoint detection, which is currently limited by predominantly singular-scale detection techniques and fails to exploit sufficient information. To tackle the aforementioned challenges, we propose an adaptive dual-stream aggregation network (ADSAN), which enables the learning of finer local representations and the acquisition of abundant spatial and semantic information by merging features from both inter-layer and intra-layer perspectives through a multi-grained approach, consolidating features within individual layers and amplifying the interaction of distinct resolution features between layers. Furthermore, our ADSAN implements the selective keypoint focus module (SKFM) algorithm to alleviate problems caused by partial occlusions and viewpoint alterations. This mechanism places greater emphasis on the most challenging keypoints, ensuring the network prioritizes and optimizes its learning around these critical points. Benefiting from the finer and more robust information of space objects extracted by the ADSAN and SKFM, our method surpasses the SOTA method PoET (5.8°, 8.1°/0.0351%, 0.0744%) by 0.5°, 0.9°, and 0.0084%, 0.0354%, achieving 5.3°, 7.2° in rotation angle errors and 0.0267%, 0.0390% in normalized translation errors on the Speed and SwissCube datasets, respectively.https://www.mdpi.com/2072-4292/16/22/41386D pose estimationspace targetkeypoint detection
spellingShingle Xu Gu
Xi Yang
Hong Liu
Dong Yang
Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space Targets
Remote Sensing
6D pose estimation
space target
keypoint detection
title Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space Targets
title_full Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space Targets
title_fullStr Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space Targets
title_full_unstemmed Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space Targets
title_short Adaptive Granularity-Fused Keypoint Detection for 6D Pose Estimation of Space Targets
title_sort adaptive granularity fused keypoint detection for 6d pose estimation of space targets
topic 6D pose estimation
space target
keypoint detection
url https://www.mdpi.com/2072-4292/16/22/4138
work_keys_str_mv AT xugu adaptivegranularityfusedkeypointdetectionfor6dposeestimationofspacetargets
AT xiyang adaptivegranularityfusedkeypointdetectionfor6dposeestimationofspacetargets
AT hongliu adaptivegranularityfusedkeypointdetectionfor6dposeestimationofspacetargets
AT dongyang adaptivegranularityfusedkeypointdetectionfor6dposeestimationofspacetargets