PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes

@article{Xiang2017PoseCNNAC,
  title={PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes},
  author={Yu Xiang and Tanner Schmidt and Venkatraman Narayanan and Dieter Fox},
  journal={ArXiv},
  year={2017},
  volume={abs/1711.00199}
}
  • Yu Xiang, Tanner Schmidt, +1 author Dieter Fox
  • Published in Robotics: Science and Systems 2017
  • Computer Science
  • Highlight Information
    Estimating the 6D pose of known objects is important for robots to interact with the real world. [...] Key Method PoseCNN estimates the 3D translation of an object by localizing its center in the image and predicting its distance from the camera. The 3D rotation of the object is estimated by regressing to a quaternion representation. We also introduce a novel loss function that enables PoseCNN to handle symmetric objects. In addition, we contribute a large scale video dataset for 6D object pose estimation named…Expand Abstract

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 266 CITATIONS

    DeepIM: Deep Iterative Matching for 6D Pose Estimation

    VIEW 28 EXCERPTS
    CITES METHODS & BACKGROUND

    End-to-end learning of local point cloud feature descriptors

    VIEW 12 EXCERPTS
    CITES BACKGROUND
    HIGHLY INFLUENCED

    Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation

    VIEW 10 EXCERPTS
    CITES RESULTS, BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Object-RPE: Dense 3D Reconstruction and Pose Estimation with Convolutional Neural Networks for Warehouse Robots

    VIEW 16 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Refining 6D Object Pose Predictions using Abstract Render-and-Compare

    VIEW 12 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    Visual Perception For Robotic Spatial Understanding

    VIEW 4 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    DeepIM: Deep Iterative Matching for 6D Pose Estimation

    VIEW 22 EXCERPTS
    CITES METHODS, BACKGROUND & RESULTS

    6D Object Pose Regression via Supervised Learning on Point Clouds

    VIEW 12 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    FILTER CITATIONS BY YEAR

    2017
    2020

    CITATION STATISTICS

    • 69 Highly Influenced Citations

    • Averaged 74 Citations per year from 2017 through 2019

    • 231% Increase in citations per year in 2019 over 2018

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 26 REFERENCES

    Learning Analysis-by-Synthesis for 6D Pose Estimation in RGB-D Images

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Uncertainty-Driven 6D Pose Estimation of Objects and Scenes from a Single RGB Image

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning 6D Object Pose Estimation Using 3D Object Coordinates

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    6-DoF object pose from semantic keypoints

    VIEW 1 EXCERPT

    DA-RNN: Semantic Mapping with Data Associated Recurrent Neural Networks

    VIEW 1 EXCERPT

    SSD-6D: Making RGB-Based 3D Detection and 6D Pose Estimation Great Again

    VIEW 1 EXCERPT

    Global Hypothesis Generation for 6D Object Pose Estimation

    VIEW 3 EXCERPTS