Reference : Leveraging Equivariant Features for Absolute Pose Regression
Scientific journals : Article
Engineering, computing & technology : Computer science
Security, Reliability and Trust
http://hdl.handle.net/10993/52233
Leveraging Equivariant Features for Absolute Pose Regression
English
Mohamed Ali, Mohamed Adel mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2 >]
Gaudilliere, Vincent mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2 >]
Ortiz Del Castillo, Miguel mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2 >]
Al Ismaeil, Kassem mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2 >]
Aouada, Djamila mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2 >]
2022
IEEE Conference on Computer Vision and Pattern Recognition.
Yes
International
[en] Computer Vision ; Deep Learning ; Pose Estimation
[en] Pose estimation enables vision-based systems to refer to their environment, supporting activities ranging from scene navigation to object manipulation. However, end-to-end approaches, that have achieved state-of-the-art performance in many perception tasks, are still unable to compete with 3D geometry-based methods in pose estimation. Indeed, absolute pose regression has been proven to be more related to image retrieval than to 3D structure. Our assumption is that statistical features learned by classical convolutional neural networks do not carry enough geometrical information for reliably solving this task. This paper studies the use of deep equivariant features for end-to-end pose regression. We further propose a translation and rotation equivariant Convolutional Neural Network whose architecture directly induces representations of camera motions into the feature space. In the context of absolute pose regression, this geometric property allows for implicitly augmenting the training data under a whole group of image plane-preserving transformations. Therefore, directly learning equivariant features efficiently compensates for learning intermediate representations that are indirectly equivariant yet data-intensive. Extensive experimental validation demonstrates that our lightweight model outperforms existing ones on standard datasets.
Interdisciplinary Centre for Security, Reliability and Trust (SnT) > Computer Vision Imaging & Machine Intelligence (CVI┬▓)
http://hdl.handle.net/10993/52233
FnR ; FNR14755859 > Djamila Aouada > MEET-A > Multi-modal Fusion Of Electro-optical Sensors For Spacecraft Pose Estimation Towards Autonomous In-orbit Operations > 01/01/2021 > 31/12/2023 > 2020

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
Musallam_Leveraging_Equivariant_Features_for_Absolute_Pose_Regression_CVPR_2022_paper.pdfPublisher postprint8.39 MBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.