Edge Enhanced Implicit Orientation Learning With Geometric Prior for 6D Pose Estimation

Wen, Yilin; Pan, Hao; Yang, Liwei; Wang, Wenping

doi:10.1109/lra.2020.3005121

Cited by 30 publications

(15 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A codebook of latent codes is created off-line, and they use a nearest neighbor search to compare a test code within the codebook. This approach is improved by [15] by adding edge priors. Our 6D pose estimation pipeline adapts the AAE concept [6] for point clouds.…”

Section: Related Workmentioning

confidence: 99%

“…Viewpoints are selected using some simple or dataset-based heuristics. One common approach is to define spheres around the textured 3D object model with fixed radii, and sample viewpoints from a hemisphere that covers the upper part of the object model [9], [16], or from the full sphere [6], [15]. The object model is rendered in the target viewpoint onto a plain or random background.…”

Section: Related Workmentioning

confidence: 99%

“…We consider three datasets, the LineMOD dataset (LM) [9], LineMOD Occlusion (LMO) [32] and YCB video dataset (YCBV) [4]. Depending on the applicability of the methods, for each dataset, we compare to a subset of stateof-the-art methods SSD-6D [29], EEPG-AAE [15], Cloud-Pose [8], PVNet [30], PoseCNN [4], DenseFusion [12], PVN3D [13] and PointVoteNet [31].…”

Section: A Datasets and Experiments Setupmentioning

confidence: 99%

“…LM [9] contains 13 objects from daily scenarios in 13 videos. Although the target objects are not occluded, this dataset is very challenging for depth based methods due to noise in the depth data [15]. LMO [32] contains 8 objects from the LM dataset.…”

Section: A Datasets and Experiments Setupmentioning

confidence: 99%

See 3 more Smart Citations

CloudAAE: Learning 6D Object Pose Regression with On-line Data Synthesis on Point Clouds

Gao

Lauri

et al. 2021

Preprint

View full text Add to dashboard Cite

It is often desired to train 6D pose estimation systems on synthetic data because manual annotation is expensive. However, due to the large domain gap between the synthetic and real images, synthesizing color images is expensive. In contrast, this domain gap is considerably smaller and easier to fill for depth information. In this work, we present a system that regresses 6D object pose from depth information represented by point clouds, and a lightweight data synthesis pipeline that creates synthetic point cloud segments for training. We use an augmented autoencoder (AAE) for learning a latent code that encodes 6D object pose information for pose regression. The data synthesis pipeline only requires texture-less 3D object models and desired viewpoints, and it is cheap in terms of both time and hardware storage. Our data synthesis process is up to three orders of magnitude faster than commonly applied approaches that render RGB image data. We show the effectiveness of our system on the LineMOD, LineMOD Occlusion, and YCB Video datasets. The implementation of our system is available at: https://github.com/GeeeG/CloudAAE.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: A Datasets and Experiments Setupmentioning

confidence: 99%

Section: A Datasets and Experiments Setupmentioning

confidence: 99%

See 2 more Smart Citations

CloudAAE: Learning 6D Object Pose Regression with On-line Data Synthesis on Point Clouds

Gao

Lauri

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…6D Pose estimation There is a massive literature of instance-level pose estimation from RGB(D) images (we refer the readers to [26] for a comprehensive survey), which can be roughly classified into three streams, i.e., by direct pose regression [50,23,2], by registering 2D and 3D points [4,36,41,33,31], and by template retrieval [18,39,40,51,48]. Most learning based methods train specialized networks for testing instances.…”

Section: Related Workmentioning

confidence: 99%

DISP6D: Disentangled Implicit Shape and Pose Learning for Scalable 6D Pose Estimation

Wen¹,

Li²,

Pan³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

6D pose estimation of rigid objects from a single RGB image has seen tremendous improvements recently by using deep learning to combat complex real-world variations, but a majority of methods build models on the per-object level, failing to scale to multiple objects simultaneously. In this paper, we present a novel approach for scalable 6D pose estimation, by self-supervised learning on synthetic data of multiple objects using a single autoencoder. To handle multiple objects and generalize to unseen objects, we disentangle the latent object shape and pose representations, so that the latent shape space models shape similarities, and the latent pose code is used for rotation retrieval by comparison with canonical rotations. To encourage shape space construction, we apply contrastive metric learning and enable the processing of unseen objects by referring to similar training objects. The different symmetries across objects induce inconsistent latent pose spaces, which we capture with a conditioned block producing shape-dependent pose codebooks by re-entangling shape and pose representations. We test our method on two multi-object benchmarks with real data, T-LESS and NOCS REAL275, and show it outperforms existing RGB-based methods in terms of pose estimation accuracy and generalization.

show abstract

Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Images

Liu

Wen

Peng

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Edge Enhanced Implicit Orientation Learning With Geometric Prior for 6D Pose Estimation

Cited by 30 publications

References 27 publications

CloudAAE: Learning 6D Object Pose Regression with On-line Data Synthesis on Point Clouds

CloudAAE: Learning 6D Object Pose Regression with On-line Data Synthesis on Point Clouds

DISP6D: Disentangled Implicit Shape and Pose Learning for Scalable 6D Pose Estimation

Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Images

Contact Info

Product

Resources

About