CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-DoF Object Pose Estimation

By Zhigang Li, Gu Wang, Xiangyang Ji. IEEE International Conference on Computer Vision, 2019

492 citations recorded by the source corpus.

6-DoF object pose estimation from a single RGB image is a fundamental and long-standing problem in computer vision. Current leading approaches solve it by training deep networks to either regress both rotation and translation from image directly or to construct 2D-3D correspondences and further solve them via PnP indirectly. We argue that rotation and translation should be treated differently for their significant difference. In this work, we propose a novel 6-DoF pose estimation approach: Coordinates-based Disentangled Pose Network (CDPN), which disentangles the pose to predict rotation and translation separately to achieve highly accurate and robust pose estimation. Our method is flexible, efficient, highly accurate and can deal with texture-less and occluded objects. Extensive experiments on LINEMOD and Occlusion datasets are conducted and demonstrate the superiority of our approach. Concretely, our approach significantly exceeds the state-of-the- art RGB-based methods on commonly used metrics.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/d1d71d83d5976e4bc49b65ab2aaa3f907482b31b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

HybridPose: 6D Object Pose Estimation Under Hybrid Representations (2020, background): “[38], BB8 [36], Pix2Pose [30], PVNet [34], CDPN [20], and DPOD [44].”
BOP Challenge 2020 on 6D Object Localization (2020, unknown): “er [10] 2019 yes - - - d icp 18 Felix&Neves-ICRA17-IET19 [46,44] 2019 yes 1/set rgb-d syn+real rgb-d icp 19 Sundermeyer-IJCV19+ICP [49] 2019 - 1/obj rgb syn+real rgb-d icp 20 Zhigang-CDPN-ICCV19 [35] 2019 - 1/obj rgb syn+real rgb - 21 PointVoteNet2 [15] 2020 - 1/obj rgb-d pbr rgb-d icp 22 Pix2Po”
GDRNPP: A Geometry-Guided and Fully Learning-Based Object Pose Estimator (2021, unknown): “To account for symmetric objects, given ¯ R , the set of all possible ground-truth rotations under symmetry, we further extend our loss to a symmetry-aware formulation L R , sym = min”
Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: a review (2020, background): “CDPN [Li et al., 2019] proposed the Coordinates-based Disentangled Pose Network (CDPN), which disentangles the pose to predict rotation and translation separately.”
FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects (2023, methodology): “The object pose is often solved by direct regression [35, 67], or constructing 2D-3D correspondences followed by P n P [47, 55], or 3D-3D correspondences followed by least squares fitting [19, 20].”
PVN3D: A Deep Point-Wise 3D Keypoints Voting Network for 6DoF Pose Estimation (2019, methodology): “They either use random forest [3, 32] or CNNs [23, 9, 27, 35, 51] to extract feature and predict the corresponding 3D object coordinates for each pixel and then vote for the final pose results.”