UDA-COPE: Unsupervised Domain Adaptation for Category-level Object Pose Estimation

Lee, Taeyeop; Lee, Byeong-Uk; Shin, Inkyu; Choe, Jaesung; Shin, Ukcheol; Kweon, In So; Yoon, Kuk-Jin

doi:10.1109/cvpr52688.2022.01447

Cited by 30 publications

(12 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address this issue, prior-based methods [21,3,30,7,17,10] leverage category-specific 3D priors (templates) to guide pose estimation. They adopt a prior-driven deformation module [30] to deform the prior for synthesizing the target object in world-space.…”

Section: Prior-based Methodsmentioning

confidence: 99%

Prior-free Category-level Pose Estimation with Implicit Space Transformation

Liu¹,

Chen²,

Ye³

et al. 2023

Preprint

View full text Add to dashboard Cite

Category-level 6D pose estimation aims to predict the poses and sizes of unseen objects from a specific category. Thanks to prior deformation, which explicitly adapts a category-specific 3D prior (i.e., a 3D template) to a given object instance, prior-based methods attained great success and have become a major research stream. However, obtaining category-specific priors requires collecting a large amount of 3D models, which is labor-consuming and often not accessible in practice. This motivates us to investigate whether priors are necessary to make prior-based methods effective. Our empirical study shows that the 3D prior itself is not the credit to the high performance. The keypoint actually is the explicit deformation process, which aligns camera and world coordinates supervised by worldspace 3D models (also called canonical space). Inspired by these observation, we introduce a simple prior-free implicit space transformation network, namely IST-Net, to transform camera-space features to world-space counterparts and build correspondence between them in an implicit manner without relying on 3D priors. Besides, we design camera-and world-space enhancers to enrich the features with pose-sensitive information and geometrical constraints, respectively. Albeit simple, IST-Net becomes the first prior-free method that achieves state-of-the-art performance, with top inference speed on the REAL275 dataset. Our code and models will be publicly available.

show abstract

Section: Prior-based Methodsmentioning

confidence: 99%

Prior-free Category-level Pose Estimation with Implicit Space Transformation

Liu¹,

Chen²,

Ye³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Other methods such as [29,30,31,32,33] predict pose and bounding box without reconstructing the full shape of the object. For the evaluation presented in this work, we limit ourselves to methods that perform both reconstruction and pose estimation, although our evaluation protocol could in principle be used for pure pose estimation methods as well.…”

Section: Related Workmentioning

confidence: 99%

“…Only methods that estimate both pose and shape and do not require an initial pose estimate are included. Therefore, pure categorial pose estimation methods such as [3,29,33,58] are excluded; tracking methods that require an initial pose estimate such as [59,60,61] are excluded; and finally, methods that explicitly rely on objects being upright such as [34,2,62] are excluded.…”

Section: Related Workmentioning

confidence: 99%

RGB-D-Based Categorical Object Pose and Shape Estimation: Methods, Datasets, and Evaluation

Bruns¹

2023

Preprint

View full text Add to dashboard Cite

Recently, various methods for 6D pose and shape estimation of objects at a per-category level have been proposed. This work provides an overview of the field in terms of methods, datasets, and evaluation protocols. First, an overview of existing works and their commonalities and differences is provided. Second, we take a critical look at the predominant evaluation protocol, including metrics and datasets. Based on the findings, we propose a new set of metrics, contribute new annotations for the Redwood dataset, and evaluate state-of-the-art methods in a fair comparison. The results indicate that existing methods do not generalize well to unconstrained orientations and are actually heavily biased towards objects being upright. We provide an easy-to-use evaluation toolbox with well-defined metrics, methods, and dataset interfaces, which allows evaluation and comparison with various state-of-the-art approaches (https://github.com/roym899/pose and shape evaluation).

show abstract

“…[2,46] gain the ability to generalize by mapping the input shape to normalized or metric-scale canonical spaces and then recovering the objects' poses via correspondence matching. Better handling of intra-category shape variation is also achieved by leveraging shape priors [4,42,56], symmetry priors [20], or domain adaptation [17,21]. Additionally, [5] enhances the perceptiveness of local geometry, and [7,55] exploit geometric consistency terms to improve the performance further.…”

Section: Introductionmentioning

confidence: 99%

“…Despite the remarkable progress of existing methods, there is still room for improvement in the performance of the category-level object pose estimation. Reconstruction and matching-based methods [17,42,46] are usually limited in speed due to the time-consuming correspondencematching procedure. Recently, various methods [5,7,20,55,56] built on 3D graph convolution (3D-GC) [23] have achieved impressive performance and run in real-time.…”

Section: Introductionmentioning

confidence: 99%

HS-Pose: Hybrid Scope Feature Extraction for Category-level Object Pose Estimation

Zheng¹,

Wang²,

Sun³

et al. 2023

Preprint

View full text Add to dashboard Cite

In this paper, we focus on the problem of category-level object pose estimation, which is challenging due to the large intra-category shape variation. 3D graph convolution (3D-GC) based methods have been widely used to extract local geometric features, but they have limitations for complex shaped objects and are sensitive to noise. Moreover, the scale and translation invariant properties of 3D-GC restrict the perception of an object's size and translation information. In this paper, we propose a simple network structure, the HS-layer, which extends 3D-GC to extract hybrid scope latent features from point cloud data for category-level object pose estimation tasks. The proposed HS-layer: 1) is able to perceive local-global geometric structure and global information, 2) is robust to noise, and 3) can encode size and translation information. Our experiments show that the simple replacement of the 3D-GC layer with the proposed HS-layer on the baseline method (GPV-Pose) achieves a significant improvement, with the performance increased by 14.5% on 5 • 2cm metric and 10.3% on IoU 75 . Our method outperforms the state-of-the-art methods by a large margin (8.3% on 5 • 2cm, 6.9% on IoU 75 ) on REAL275 dataset and runs in real-time (50 FPS) 1 .

show abstract

UDA-COPE: Unsupervised Domain Adaptation for Category-level Object Pose Estimation

Cited by 30 publications

References 54 publications

Prior-free Category-level Pose Estimation with Implicit Space Transformation

Prior-free Category-level Pose Estimation with Implicit Space Transformation

RGB-D-Based Categorical Object Pose and Shape Estimation: Methods, Datasets, and Evaluation

HS-Pose: Hybrid Scope Feature Extraction for Category-level Object Pose Estimation

Contact Info

Product

Resources

About