ZebraPose: Coarse to Fine Surface Encoding for 6DoF Object Pose Estimation (2022-03-17T00:00:00.000000Z)

TL;DR

This work presents a discrete descriptor, which can represent the object surface densely by incorporating a hierarchical binary grouping, and proposes a coarse to fine training strategy, which enables fine-grained correspondence prediction of the 6DoF pose.

Abstract

Establishing correspondences from image to 3D has been a key task of 6DoF object pose estimation for a long time. To predict pose more accurately, deeply learned dense maps replaced sparse templates. Dense methods also improved pose estimation in the presence of occlusion. More recently researchers have shown improvements by learning object fragments as segmentation. In this work, we present a discrete descriptor, which can represent the object surface densely. By incorporating a hierarchical binary grouping, we can encode the object surface very efficiently. Moreover, we propose a coarse to fine training strategy, which enables fine-grained correspondence prediction. Finally, by matching predicted codes with object surface and using a PnP solver, we estimate the 6DoF pose. Results on the public LM-O and YCB-V datasets show major improvement over the state of the art w.r.t. ADD(-S) metric, even surpassing RGB-D based methods in some cases.

Authors

N. Navab

26 papers

D. Stricker

6 papers

Yongzhi Su

2 papers

TL;DR

Abstract

Authors

References74 items

PR-GCN: A Deep Graph Convolutional Network with Point Refinement for 6D Pose Estimation

SO-Pose: Exploiting Self-Occlusion for Direct 6D Pose Estimation

RePOSE: Fast 6D Object Pose Refinement via Deep Texture Rendering

FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation

SynPo-Net—Accurate and Fast CNN-Based 6DoF Object Pose Estimation Using Synthetic Training

I Like to Move It: 6D Pose Estimation as an Action Decision Process

CosyPose: Consistent multi-view multi-object 6D pose estimation

ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual–Inertial, and Multimap SLAM

Self6D: Self-Supervised Monocular 6D Object Pose Estimation

EPOS: Estimating 6D Pose of Objects With Symmetries

Introducing Pose Consistency and Warp-Alignment for Self-Supervised 6D Object Pose Estimation in Color Images

HybridPose: 6D Object Pose Estimation Under Hybrid Representations

Category-Level Articulated Object Pose Estimation

Single-Stage 6D Object Pose Estimation

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

Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation

Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer

Progressive-X: Efficient, Anytime, Multi-Model Fitting Algorithm

Neural-Guided RANSAC: Learning Where to Sample Model Hypotheses

DeceptionNet: Network-Driven Domain Randomization

FCOS: Fully Convolutional One-Stage Object Detection

DPOD: 6D Pose Object Detector and Refiner

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

PVNet: Pixel-Wise Voting Network for 6DoF Pose Estimation

On the Continuity of Rotation Representations in Neural Networks

Segmentation-Driven 6D Object Pose Estimation

Explaining the Ambiguity of Object Detection and 6D Pose From Visual Data

Dealing with Ambiguity in Robotic Grasping via Multiple Predictions

Learning 6DoF Object Poses from Synthetic Single Channel Images

Deep Model-Based 6D Pose Refinement in RGB

Implicit 3D Orientation Learning for 6D Object Detection from RGB Images

A Survey on 3D Cameras: Metrological Comparison of Time-of-Flight, Structured-Light and Active Stereoscopy Technologies

Making Deep Heatmaps Robust to Partial Occlusions for 3D Object Pose Estimation

6D pose estimation using an improved method based on point pair features

6DoF Object Tracking based on 3D Scans for Augmented Reality Remote Live Support

Learning Less is More - 6D Camera Localization via 3D Surface Regression

Neural 3D Mesh Renderer

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

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

Rethinking Atrous Convolution for Semantic Image Segmentation

Camera Pose Filtering with Local Regression Geodesics on the Riemannian Manifold of Dual Quaternions

BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects without Using Depth

Pose Estimation for Augmented Reality: A Hands-On Survey

DSAC — Differentiable RANSAC for Camera Localization

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

Robot Guidance Using Machine Vision Techniques in Industrial Environments: A Comparative Review

Deep Residual Learning for Image Recognition

A Stereo Vision Approach for Cooperative Robotic Movement Therapy

The YCB object and Model set: Towards common benchmarks for manipulation research

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization

Learning descriptors for object recognition and 3D pose estimation

Adam: A Method for Stochastic Optimization

Learning 6D Object Pose Estimation Using 3D Object Coordinates

OpenDR: An Approximate Differentiable Renderer

Model Based Training, Detection and Pose Estimation of Texture-Less 3D Objects in Heavily Cluttered Scenes

General Midpoint Subdivision

Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes

ORB: An efficient alternative to SIFT or SURF

The MOPED framework: Object recognition and pose estimation for manipulation

Fast Approximate Energy Minimization with Label Costs

k-means++: the advantages of careful seeding

Distinctive Image Features from Scale-Invariant Keypoints

Pattern codification strategies in structured light systems

A Method of Time-Coded Parallel Planes of Light for Depth Measurement

LieNet: Real-time Monocular Object Instance 6D Pose Estimation

Semantic Image Segmentation

Distinctive Image Features from Scale-Invariant Keypoints Abstract by Matthijs Dorst Based on the paper by

Error Model of a Coded-Light Range Sensor

Pose Estimation for

Surface measurement by space-encoded projected beam systems

Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Gradient Response Maps for Real-time Detection of Texture-less Objects

Noname manuscript No. (will be inserted by the editor) EPnP: An Accurate O(n) Solution to the PnP Problem

ORB: an efﬁcient alternative to SIFT or SURF

Field of Study

Journal Information