computer-vision-7

6D Pose Estimation

3260 papers • 126 benchmarks • 313 datasets

Image: Zeng et al

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in 6d-pose-estimation-7

Trend

Dataset

Best Model

Actions

YCB-Video

LineMOD

OPT

Libraries

i

Use these libraries to find 6d-pose-estimation-7 models and implementations

thodan/bop_toolkit

3 papers 400

Datasets

LM

Subtasks

hand-object pose Robot Pose Estimation

Most implemented papers

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

L. Guibas, Shuran Song, Srinath Sridhar, He Wang, Jingwei Huang, Julien P. C. Valentin•Tue Jan 08 2019

The proposed method is able to robustly estimate the pose and size of unseen object instances in real environments while also achieving state-of-the-art performance on standard 6D pose estimation benchmarks.

819

Content

ApolloCar3D

3D-BSLS-6D

dlr-rm/3dobjecttracking

2 papers 708

wenbowen123/BundleTrack

2 papers 604

NERDS 360

0

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

D. Fox, Yu Xiang, Tanner Schmidt, V. Narayanan•Tue Oct 31 2017

This work introduces PoseCNN, a new Convolutional Neural Network for 6D object pose estimation, which is highly robust to occlusions, can handle symmetric objects, and provide accurate pose estimation using only color images as input.

2186 0

Paper Graph

DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusion

S. Savarese, Li Fei-Fei, Yuke Zhu, Roberto Martín-Martín, Chen Wang, Danfei Xu, Cewu Lu•Mon Jan 14 2019

DenseFusion is a generic framework for estimating 6D pose of a set of known objects from RGB-D images that processes the two data sources individually and uses a novel dense fusion network to extract pixel-wise dense feature embedding, from which the pose is estimated.

1092 0

Paper Graph

Estimating 6D Pose From Localizing Designated Surface Keypoints

Haoshu Fang, Cewu Lu, Zelin Zhao, Gao Peng, Haoyu Wang, Chengkun Li•Mon Dec 03 2018

The core of the approach is that a set of surface points on target object model are designated as keypoints and then train a keypoint detector (KPD) to localize them and a PnP algorithm can recover the 6D pose according to the 2D-3D relationship of keypoints.

34 0

Paper Graph

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

T. Patten, M. Vincze, Kiru Park•Mon Aug 19 2019

A novel pose estimation method that predicts the 3D coordinates of each object pixel without textured models, and a novel loss function, the transformer loss, is proposed to handle symmetric objects by guiding predictions to the closest symmetric pose.

514 0

Paper Graph

Segmentation-Driven 6D Object Pose Estimation

P. Fua, M. Salzmann, Yinlin Hu, Joachim Hugonot•Wed Dec 05 2018

This paper introduces a segmentation-driven 6D pose estimation framework where each visible part of the objects contributes a local pose prediction in the form of 2D keypoint locations and uses a predicted measure of confidence to combine these pose candidates into a robust set of 3D-to-2D correspondences.

304 0

Paper Graph

BOP Challenge 2020 on 6D Object Localization

Eric Brachmann, Frank Michel, C. Rother, Jiri Matas, Tomás Hodan, M. Sundermeyer, Bertram Drost, Yann Labbé•Mon Sep 14 2020

Although the top-performing methods rely on RGB-D image channels, strong results were achieved when only RGB channels were used at both training and test time, and the photorealism of PBR images was demonstrated effective despite the augmentation.

312 0

Paper Graph

PVN3D: A Deep Point-Wise 3D Keypoints Voting Network for 6DoF Pose Estimation

Yisheng He, Wei Sun, Haibin Huang, Haoqiang Fan, Jian Sun, Jianran Liu•Sun Nov 10 2019

A deep Hough voting network is proposed to detect 3D keypoints of objects and then estimate the 6D pose parameters within a least-squares fitting manner, which is a natural extension of 2D-keypoint approaches that successfully work on RGB based 6DoF estimation.

547 0

Paper Graph

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

Josef Sivic, Mathieu Aubry, Yann Labb'e, Justin Carpentier•Tue Aug 18 2020

The proposed method, dubbed CosyPose, outperforms current state-of-the-art results for single-view and multi-view 6D object pose estimation by a large margin on two challenging benchmarks: the YCB-Video and T-LESS datasets.

500 0

Paper Graph

EfficientPose: An efficient, accurate and scalable end-to-end 6D multi object pose estimation approach

Yannick Bukschat, Marcus Vetter•Sun Nov 08 2020

EfficientPose is a new approach for 6D object pose estimation that achieves a new state-of-the-art accuracy of 97.35% in terms of the ADD(-S) metric on the widely-used 6D pose estimation benchmark dataset Linemod using RGB input, while still running end-to-end at over 27 FPS.

127 0

Paper Graph

Adding a benchmark result helps the community track progress.