End-to-End 6DoF Pose Estimation From Monocular RGB Images

Zou, Wenbin; Wu, Di; Tian, Shishun; Xiang, Canqun; Li, Xia; Zhang, Lu

doi:10.1109/tce.2021.3057137

Cited by 18 publications

(9 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, quaternion is a popular representation. It is utilized such as in PoseCNN [1], PoseNet [10], and 6D-VNet [11]. Another 4-dimensional representation is an axis-rotation representation, introduced such as in MapNet [12].…”

Section: A Continuous Rotation Representationsmentioning

confidence: 99%

Probabilistic Rotation Representation With an Efficiently Computable Bingham Loss Function and Its Application to Pose Estimation

Sato¹,

Ikeda²,

Nishiwaki³

2022

Preprint

View full text Add to dashboard Cite

In recent years, a deep learning framework has been widely used for object pose estimation. While quaternion is a common choice for rotation representation of 6D pose, it cannot represent an uncertainty of the observation. In order to handle the uncertainty, Bingham distribution is one promising solution because this has suitable features, such as a smooth representation over SO(3), in addition to the ambiguity representation. However it requires the complex computation of the normalizing constants. This is the bottleneck of loss computation in training neural networks based on Bingham representation. As such, we propose a fast-computable and easy-to-implement loss function for Bingham distribution. We also show not only to examine the parametrization of Bingham distribution but also an application based on our loss function.

show abstract

Section: A Continuous Rotation Representationsmentioning

confidence: 99%

Probabilistic Rotation Representation With an Efficiently Computable Bingham Loss Function and Its Application to Pose Estimation

Sato¹,

Ikeda²,

Nishiwaki³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The single end-to-end network was also used to tackle 6DoF pose estimation from single monocular RGB images [29,30]. Bayesian PoseNet was presented to achieve end-to-end train, and the uncertainty of repositioning was obtained through the posterior distribution from Bayesian CNN weights.…”

Section: Related Workmentioning

confidence: 99%

“…Bayesian PoseNet was presented to achieve end‐to‐end train, and the uncertainty of repositioning was obtained through the posterior distribution from Bayesian CNN weights. Both PoseNet [27] and Bayesian PoseNet models [30] showed great performance in indoor and outdoor relocation environment. The work in [27] was extended to address the difficulty in setting the super parameters of the loss function in the PoseNet by presenting a more basic theoretical treatment for, and explored many new loss functions for learning the camera attitude based on geometry and scene re‐projection errors.…”

Section: Introductionmentioning

confidence: 99%

Spaceborne pose determination based on image to 3D digital surface model matching

Xing

Gao

Chen

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Developing an accurate onboard‐camera pose estimation is one major challenge of satellite systems, and the attempt of improving remote sensing camera pose accuracy never ceases. The camera pose can be recovered by aligning a captured 2D image and a 3D digital surface model of the corresponding scene. In this paper, a novel camera pose estimation method from captured images with the over known real scene 3D products is proposed to enhance remote sensing camera attitude accuracy. The purpose of this estimation is to determine the pose of a camera purely from an image based on a known 3D model, where 3D products of very high spatial resolution are projected onto image space by virtual camera system with error‐contained initial exterior orientation parameters, and whether the pose of the camera can be determined precisely depends on the 2D–3D registration result. The process consists of two steps: (1) feature extraction and (2) similarity measure and registration. Furthermore, the proposed method revises the rotation matrix and translation vector by utilizing formulation based on quaternion representation of rotation, respectively. We evaluate our method on challenging simulation data and results show that acceptable accuracy of camera pose can be achieved.

show abstract

“…Deep learning technology has achieved considerably good performance in computer vision, natural language processing, speech recognition, and other fields. In recent years, ModelNet (Wu et al, 2015 ), ShapeNet (Yang et al, 2021 ), ScanNet (Zou et al, 2021 ), and other publicly available datasets have also driven research in 3D model classification based on deep learning. 3D model classification methods based on deep learning can be divided into three categories based on the representation of the input data: voxel-based, point cloud-based, and multi-view-based.…”

Section: Introductionmentioning

confidence: 99%

Multi-view SoftPool attention convolutional networks for 3D model classification

Wang

Chen

et al. 2022

Front. Neurorobot.

View full text Add to dashboard Cite

IntroductionExisting multi-view-based 3D model classification methods have the problems of insufficient view refinement feature extraction and poor generalization ability of the network model, which makes it difficult to further improve the classification accuracy. To this end, this paper proposes a multi-view SoftPool attention convolutional network for 3D model classification tasks.MethodsThis method extracts multi-view features through ResNest and adaptive pooling modules, and the extracted features can better represent 3D models. Then, the results of the multi-view feature extraction processed using SoftPool are used as the Query for the self-attentive calculation, which enables the subsequent refinement extraction. We then input the attention scores calculated by Query and Key in the self-attention calculation into the mobile inverted bottleneck convolution, which effectively improves the generalization of the network model. Based on our proposed method, a compact 3D global descriptor is finally generated, achieving a high-accuracy 3D model classification performance.ResultsExperimental results showed that our method achieves 96.96% OA and 95.68% AA on ModelNet40 and 98.57% OA and 98.42% AA on ModelNet10.DiscussionCompared with a multitude of popular methods, our algorithm model achieves the state-of-the-art classification accuracy.

show abstract

End-to-End 6DoF Pose Estimation From Monocular RGB Images

Cited by 18 publications

References 45 publications

Probabilistic Rotation Representation With an Efficiently Computable Bingham Loss Function and Its Application to Pose Estimation

Probabilistic Rotation Representation With an Efficiently Computable Bingham Loss Function and Its Application to Pose Estimation

Spaceborne pose determination based on image to 3D digital surface model matching

Multi-view SoftPool attention convolutional networks for 3D model classification

Contact Info

Product

Resources

About