mlsanityData to 14 Oct 2026

Robust 6DoF Pose Estimation Against Depth Noise and a Comprehensive Evaluation on a Mobile Dataset

arXiv 2309.13570 ↗First posted 24 Sep 2023cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

Robust 6DoF pose estimation with mobile devices is the foundation for applications in robotics, augmented reality, and digital twin localization. In this paper, we extensively investigate the robustness of existing RGBD-based 6DoF pose estimation methods against varying levels of depth sensor noise. We highlight that existing 6DoF pose estimation methods suffer significant performance discrepancies due to depth measurement inaccuracies. In response to the robustness issue, we present a simple and effective transformer-based 6DoF pose estimation approach called DTTDNet, featuring a novel geometric feature filtering module and a Chamfer distance loss for training. Moreover, we advance the field of robust 6DoF pose estimation and introduce a new dataset -- Digital Twin Tracking Dataset Mobile (DTTD-Mobile), tailored for digital twin object tracking with noisy depth data from the mobile RGBD sensor suite of the Apple iPhone 14 Pro. Extensive experiments demonstrate that DTTDNet significantly outperforms state-of-the-art methods at least 4.32, up to 60.74 points in ADD metrics on the DTTD-Mobile. More importantly, our approach exhibits superior robustness to varying levels of measurement noise, setting a new benchmark for robustness to measurement noise. The project page is publicly available at https://openark-berkeley.github.io/DTTDNet/.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 42 of its 65 references that are in the map. First posted on arXiv 24 Sep 2023.

Builds on

42 of 65 references.

References · 42

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  4. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  5. 2016Pyramid Scene Parsing Networkcited by 2,479
  6. 2021ViLT: Vision-and-Language Transformer Without Convolution or Region Supervisioncited by 1,230
  7. 2020ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAMcited by 878
  8. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 775
  9. 2021ARKitScenes: A Diverse Real-World Dataset For 3D Indoor Scene Understanding Using Mobile RGB-D Datacited by 454
  10. 2018PVNet: Pixel-wise Voting Network for 6DoF Pose Estimationcited by 385
  11. 2019DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusioncited by 355
  12. 2020CosyPose: Consistent multi-view multi-object 6D pose estimationcited by 287
30 more
  1. 2017T-LESS: An RGB-D Dataset for 6D Pose Estimation of Texture-less Objectscited by 210
  2. 2021Vision Transformers are Robust Learnerscited by 207
  3. 2019PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimationcited by 204
  4. 2022MegaPose: 6D Pose Estimation of Novel Objects via Render & Comparecited by 201
  5. 2019DPOD: 6D Pose Object Detector and Refinercited by 189
  6. 2019Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Drivingcited by 188
  7. 2022OnePose: One-Shot Object Pose Estimation without CAD Modelscited by 157
  8. 2023BundleSDF: Neural 6-DoF Tracking and 3D Reconstruction of Unknown Objectscited by 148
  9. 2021FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimationcited by 144
  10. 2022Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Imagescited by 121
  11. 2019Monocular 3D Object Detection with Pseudo-LiDAR Point Cloudcited by 113
  12. 2016Vehicle Detection from 3D Lidar Using Fully Convolutional Networkcited by 107
  13. 2017Intel RealSense Stereoscopic Depth Camerascited by 83
  14. 2022FS6D: Few-Shot 6D Pose Estimation of Novel Objectscited by 81
  15. 2019KeyPose: Multi-View 3D Labeling and Keypoint Estimation for Transparent Objectscited by 75
  16. 2021Towards Fast and Accurate Real-World Depth Super-Resolution: Benchmark Dataset and Baselinecited by 61
  17. 2019Total Denoising: Unsupervised Learning of 3D Point Cloud Cleaningcited by 58
  18. 2022Templates for 3D Object Pose Estimation Revisited: Generalization to New Objects and Robustness to Occlusionscited by 54
  19. 2023Mimetic Initialization of Self-Attention Layerscited by 54
  20. 2022OVE6D: Object Viewpoint Encoding for Depth-based 6D Object Pose Estimationcited by 43
  21. 2021StereOBJ-1M: Large-scale Stereo Image Dataset for 6D Object Pose Estimationcited by 40
  22. 2017LabelFusion: A Pipeline for Generating Ground Truth Labels for Real RGBD Data of Cluttered Scenescited by 32
  23. 2022Uni6D: A Unified CNN Framework without Projection Breakdown for 6D Pose Estimationcited by 22
  24. 20193D Point Cloud Denoising via Deep Neural Network based Local Surface Estimationcited by 20
  25. 2022ES6D: A Computation Efficient and Symmetry-Aware 6D Pose Regression Frameworkcited by 19
  26. 2023MobileBrick: Building LEGO for 3D Reconstruction on Mobile Devicescited by 15
  27. 2023ARKitTrack: A New Diverse Dataset for Tracking Using Mobile RGB-D Datacited by 12
  28. 2023Token Boosting for Robust Self-Supervised Visual Transformer Pre-trainingcited by 4
  29. 2023Digital Twin Tracking Dataset (DTTD): A New RGB+Depth 3D Dataset for Longer-Range Object Tracking Applicationscited by 3
  30. 2023Enhancing GAN-Based Vocoders with Contrastive Learning Under Data-limited Conditioncited by 2

Built on by

Cited by year

Citing papers per year: 2023: 1.

1 paper.

  1. 2023Jenga Stacking Based on 6D Pose Estimation for Architectural Form Finding Processcited by 0