Synthetic Ground Truth for Presegmentation of Known Objects for Effortless Pose Estimation

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

48 Downloads (Pure)

Abstract

We present a method for generating synthetic ground truth for training segmentation networks for presegmenting point clouds in pose estimation problems. Our method replaces global pose estimation algorithms such as RANSAC which requires manual fine-tuning with a robust CNN, without having to hand-label segmentation masks for the given object. The data is generated by blending cropped images of the objects with arbitrary backgrounds. We test the method in two scenarios, and show that networks trained on the generated data segments the objects with high accuracy, allowing them to be used in a pose estimation pipeline.
Original languageEnglish
Title of host publicationProceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP
EditorsGiovanni Maria Farinella, Petia Radeva, Jose Braz
Volume4
PublisherSCITEPRESS Digital Library
Publication date2020
Pages482-489
ISBN (Electronic) 978-989-758-402-2
DOIs
Publication statusPublished - 2020
Event15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Valletta, Malta
Duration: 27. Feb 202029. Feb 2020
Conference number: 15

Conference

Conference15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Number15
CountryMalta
CityValletta
Period27/02/202029/02/2020
SeriesVISIGRAPP 2020 - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Volume4

Keywords

  • Object Segmentation
  • Pose Estimation
  • Synthetic Ground Truth Generation

Fingerprint Dive into the research topics of 'Synthetic Ground Truth for Presegmentation of Known Objects for Effortless Pose Estimation'. Together they form a unique fingerprint.

  • Related projects

    Cite this

    Haarslev, F., Juel, W. K., Krüger, N., & Bodenhagen, L. (2020). Synthetic Ground Truth for Presegmentation of Known Objects for Effortless Pose Estimation. In G. M. Farinella, P. Radeva, & J. Braz (Eds.), Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP (Vol. 4, pp. 482-489). SCITEPRESS Digital Library. VISIGRAPP 2020 - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Vol.. 4 https://doi.org/10.5220/0009163904820489