Accurate estimation of the 6D pose of an object (3D translation and 3D rotation) from RGB/RGB-D images is of great importance to many higher-level tasks such as robotic manipulation, augmented reality, and autonomous driving.
The introduction of RGB-D sensors, advent of deep learning, and novel data generation pipelines led to substantial improvements in 6D object pose estimation. Yet there remain challenges to address such as robustness against severe occlusion and clutter, scalability to multiple objects, and fast and reliable object learning/modelling. As collecting ground-truth object poses is often prohibitively expensive, effective techniques for synthesis of training data and for synthetic-to-real domain transfer are in high demand. Despite all the progress, the reliability and execution speed required in industrial settings has not been reached.
Besides 6D pose estimation of specific rigid objects, the workshop focuses on other related topics such as pose estimation of deformable or articulated objects, object categories, and relative pose estimation without 3D models.
In this workshop, people working on relevant topics in both academia and industry will share up-to-date advances and identify open problems. The workshop will feature several high-quality invited talks, presentations of accepted workshop papers, and presents the results of the BOP Challenge 2022 on 6D object pose estimation.
We invite submissions to the BOP Challenge 2022 on 6D object localization, object detection and segmentation. While the 6D object localization task follows the same evaluation methodology as in the 2019 and 2020 editions of the challenge, this year we additionally evaluate 2D object detection and segmentation tasks, which precede many recent 6D pose estimation methods. Evaluating the detection/segmentation stage and the pose estimation stage separately will help us to better understand advances in the two stages.
Awards will be presented to the top-performing methods in several categories during the R6D workshop at ECCV 2022.
We invite paper submissions about unpublished work. If accepted, the papers will be published in the ECCV workshop proceedings and presented at the workshop. The papers must follow the format of the main conference (LaTeX template) and be submitted to CMT.
The covered topics include but are not limited to:
Paper submission deadline: August 1, 2022 (11:59PM PST; no extensions)
Paper acceptance notification: August 15, 2022
Paper camera-ready version: August 22, 2022 (11:59PM PST)
Deadline for submissions to the BOP Challenge 2022: October TBD, 2022 (11:59PM UTC)
Workshop date: October TBD, 2022
Martin Sundermeyer, DLR German Aerospace Center, firstname.lastname@example.org
Tomáš Hodaň, Reality Labs at Meta, email@example.com
Yann Labbé, Inria Paris
Gu Wang, Tsinghua University
Lingni Ma, Reality Labs at Meta
Eric Brachmann, Niantic
Bertram Drost, MVTec
Rigas Kouskouridas, Scape Technologies
Ales Leonardis, University of Birmingham
Carsten Steger, Technical University of Munich, MVTec
Vincent Lepetit, ENPC ParisTech, Technical University Graz
Jiří Matas, Czech Technical University in Prague