Semi-automatic 3D Object Keypoint Annotation and Detection for the Masses

被引:2
|
作者
Blomqvist, Kenneth [1 ]
Chung, Jen Jen [1 ]
Ott, Lionel [1 ]
Siegwart, Roland [1 ]
机构
[1] Swiss Fed Inst Technol, Autonomous Syst Lab, Zurich, Switzerland
基金
欧盟地平线“2020”;
关键词
D O I
10.1109/ICPR56361.2022.9956263
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Creating computer vision datasets requires careful planning and lots of time and effort. In robotics research, we often have to use standardized objects, such as the YCB object set, for tasks such as object tracking, pose estimation, grasping and manipulation, as there are datasets and pre-learned methods available for these objects. This limits the impact of our research since learning-based computer vision methods can only be used in scenarios that are supported by existing datasets. In this work, we present a full object keypoint tracking toolkit, encompassing the entire process from data collection, labeling, model learning and evaluation. We present a semi-automatic way of collecting and labeling datasets using a wrist mounted camera on a standard robotic arm. Using our toolkit and method, we are able to obtain a working 3D object keypoint detector and go through the whole process of data collection, annotation and learning in just a couple hours of active time.
引用
收藏
页码:3908 / 3914
页数:7
相关论文
共 50 条
  • [41] An interactive tool for manual, semi-automatic and automatic video annotation
    Bianco, Simone
    Ciocca, Gianluigi
    Napoletano, Paolo
    Schettini, Raimondo
    COMPUTER VISION AND IMAGE UNDERSTANDING, 2015, 131 : 88 - 99
  • [42] Attentional Keypoint Detection on Point Clouds for 3D Object Part Segmentation
    Zhou, Feng
    Zhang, Qi
    Zhu, He
    Liu, Shibo
    Jiang, Na
    Cai, Xingquan
    Qi, Qianfang
    Hu, Yong
    APPLIED SCIENCES-BASEL, 2023, 13 (23):
  • [43] 3D BAT: A Semi-Automatic, Web-based 3D Annotation Toolbox for Full-Surround, Multi-Modal Data Streams
    Zimmer, Walter
    Rangesh, Akshay
    Trivedi, Mohan
    2019 30TH IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV19), 2019, : 1816 - 1821
  • [44] Automatic Pseudo-LiDAR Annotation: Generation of Training Data for 3D Object Detection Networks
    Oh, Changsuk
    Jang, Youngseok
    Shim, Dongseok
    Kim, Changhyeon
    Kim, Junha
    Kim, H. Jin
    IEEE ACCESS, 2024, 12 : 14227 - 14237
  • [45] Improving Deep Learning by Incorporating Semi-automatic Moving Object Annotation and Filtering for Vision-based Vehicle Detection
    Feng, Ri-Chen
    Lin, Daw-Tung
    Chen, Ken-Min
    Lin, Yi-Yao
    Liu, Chin-De
    2019 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC), 2019, : 2484 - 2489
  • [46] A novel method for semi-automatic 2D to 3D video conversion
    Wu, Chenglei
    Er, Guihua
    Xie, Xudong
    Li, Tao
    Cao, Xun
    Dai, Qionghai
    2008 3DTV-CONFERENCE: THE TRUE VISION - CAPTURE, TRANSMISSION AND DISPLAY OF 3D VIDEO, 2008, : 45 - 48
  • [47] A Simple Semi-automatic Technique for 2D to 3D Video Conversion
    Chen, Jun
    Zhao, Jianhui
    Wang, Xiaomao
    Huang, Chuanhe
    Dong, Erqian
    Chen, Bingyu
    Yuan, Zhiyong
    ARTIFICIAL INTELLIGENCE AND COMPUTATIONAL INTELLIGENCE, PT II, 2011, 7003 : 336 - 343
  • [48] EasyLabel: A Semi-Automatic Pixel-wise Object Annotation Tool for Creating Robotic RGB-D Datasets
    Suchi, Markus
    Patten, Timothy
    Fischinger, David
    Vincze, Markus
    2019 INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2019, : 6678 - 6684
  • [49] Semi-automatic Learning Framework Combining Object Detection and Background Subtraction
    Alejandro, Sugino Nicolas
    Minematsu, Tsubasa
    Shimada, Atsushi
    Shibata, Takashi
    Taniguchi, Rin-ichiro
    Kaneko, Eiji
    Miyano, Hiroyoshi
    PROCEEDINGS OF THE 15TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS, VOL 5: VISAPP, 2020, : 96 - 106
  • [50] Automatic object annotation in streamed and remotely explored large 3D reconstructions
    Benjamin H?ller
    Annette Mossel
    Hannes Kaufmann
    ComputationalVisualMedia, 2021, 7 (01) : 71 - 86