Detect Globally, Label Locally: Learning Accurate 6-DOF Object Pose Estimation by Joint Segmentation and Coordinate Regression

@article{Nigam2018DetectGL,
  title={Detect Globally, Label Locally: Learning Accurate 6-DOF Object Pose Estimation by Joint Segmentation and Coordinate Regression},
  author={Apurv Nigam and Adrian Penate-Sanchez and L. Agapito},
  journal={IEEE Robotics and Automation Letters},
  year={2018},
  volume={3},
  pages={3960-3967}
}
  • Apurv Nigam, Adrian Penate-Sanchez, L. Agapito
  • Published 2018
  • Engineering, Computer Science
  • IEEE Robotics and Automation Letters
  • Coordinate regression has established itself as one of the most successful current trends in model-based 6 degree of freedom (6-DOF) object pose estimation from a single image. The underlying idea is to train a system that can regress the three-dimensional coordinates of an object, given an input RGB or RGB-D image and known object geometry, followed by a robust procedure such as RANSAC to optimize the object pose. These coordinate regression based approaches exhibit state-of-the-art… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-DoF Object Pose Estimation
    • 35
    • Highly Influenced
    • Open Access
    EPOS: Estimating 6D Pose of Objects With Symmetries
    • 5
    • Open Access
    DeepHMap++: Combined Projection Grouping and Correspondence Learning for Full DoF Pose Estimation
    • 4
    • Highly Influenced
    • Open Access
    Memory Segment Matching Network Based Image Geo-Localization

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 36 REFERENCES
    ImageNet classification with deep convolutional neural networks
    • 51,806
    • Open Access
    ImageNet Classification with Deep Convolutional Neural Networks
    • 51,806
    • Open Access
    A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography
    • 8,095
    • Highly Influential
    • Open Access
    Model Based Training, Detection and Pose Estimation of Texture-Less 3D Objects in Heavily Cluttered Scenes
    • 519
    • Highly Influential
    • Open Access
    VoxNet: A 3D Convolutional Neural Network for real-time object recognition
    • 1,310
    • Open Access
    Uncertainty-Driven 6D Pose Estimation of Objects and Scenes from a Single RGB Image
    • 242
    • Highly Influential
    • Open Access
    Learning 6D Object Pose Estimation Using 3D Object Coordinates
    • 308
    • Highly Influential
    • Open Access
    Scene Coordinate Regression Forests for Camera Relocalization in RGB-D Images
    • 354
    • Highly Influential
    • Open Access
    BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects without Using Depth
    • 258
    • Highly Influential
    • Open Access