mlsanityData to 14 Oct 2026

Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains

arXiv 2105.14391 ↗First posted 29 May 2021cs.CV cs.ROcited by 5

Not on the map. Why?

Abstract (arXiv)

Tracking the 6D pose of objects in video sequences is important for robot manipulation. This work presents se(3)-TrackNet, a data-driven optimization approach for long term, 6D pose tracking. It aims to identify the optimal relative pose given the current RGB-D observation and a synthetic image conditioned on the previous best estimate and the object's model. The key contribution in this context is a novel neural network architecture, which appropriately disentangles the feature encoding to help reduce domain shift, and an effective 3D orientation representation via Lie Algebra. Consequently, even when the network is trained solely with synthetic data can work effectively over real images. Comprehensive experiments over multiple benchmarks show se(3)-TrackNet achieves consistently robust estimates and outperforms alternatives, even though they have been trained with real images. The approach runs in real time at 90.9Hz. Code, data and supplementary video for this project are available at https://github.com/wenbowen123/iros20-6d-pose-tracking

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 18 of its 32 references that are in the map. First posted on arXiv 29 May 2021.

Builds on

18 of 32 references.

References · 18

  1. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  2. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,059
  3. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 775
  4. 2016Direct Sparse Odometrycited by 485
  5. 2019DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusioncited by 355
  6. 2018DeepIM: Deep Iterative Matching for 6D Pose Estimationcited by 283
  7. 2018Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objectscited by 241
  8. 2019PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimationcited by 204
  9. 2019Implicit 3D Orientation Learning for 6D Object Detection from RGB Imagescited by 170
  10. 2019PoseRBPF: A Rao-Blackwellized Particle Filter for 6D Object Pose Trackingcited by 91
  11. 2020se(3)-TrackNet: Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domainscited by 90
  12. 2020Robust, Occlusion-aware Pose Estimation for Objects Grasped by Adaptive Handscited by 26
6 more
  1. 2016Depth-Based Object Tracking Using a Robust Gaussian Filtercited by 22
  2. 2015Probabilistic Object Tracking using a Range Cameracited by 21
  3. 2018A Region-based Gauss-Newton Approach to Real-Time Monocular Multiple Object Trackingcited by 21
  4. 2017Deep 6-DOF Trackingcited by 13
  5. 2020Task-driven Perception and Manipulation for Constrained Placement of Unknown Objectscited by 12
  6. 2019Scene-level Pose Estimation for Multiple Instances of Densely Packed Objectscited by 8

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 3; 2023: 1.

5 papers.

  1. 2022You Only Demonstrate Once: Category-Level Manipulation from Single Visual Demonstrationcited by 81
  2. 2023TTA-COPE: Test-Time Adaptation for Category-Level Object Pose Estimationcited by 32
  3. 2021SRT3D: A Sparse Region-Based 3D Object Tracking Approach for the Real Worldcited by 19
  4. 2022Accidental Turntables: Learning 3D Pose by Watching Objects Turncited by 1
  5. 2022Few-Shot Object Detection in Unseen Domainscited by 0