Paper PLMP – Point-Line Minimal Problems in Complete Multi-View Visibility co-authored by Tomas Pajdla from Applied Algebra and Geometry Group of the CIIRC CTU in Prague won the Best Student Paper Award at the IEEE/CVF International Conference on Computer Vision 2019 (ICCV 2019) that took place in Seoul from October 29 until November 1, 2019. ICCV is the premier bi-annual international computer vision conference.
The paper presents a complete classification of all minimal problems for generic arrangements of points and lines fully observed by calibrated perspective cameras. This theoretical result makes an important contribution to understanding the space of geometrical problems in 3D reconstruction that can be efficiently solved. It is based on techniques from nonlinear algebra, such as basic algebraic geometry, symbolic elimination using Groebner bases computation, and numerical solving of polynomial systems using Homotopy Continuation and Monodromy.
This is an important result as it definitely finalizes the understanding of certain class of geometrical problems that are important in computer vision.
In applications, this classification of minimal problems opens-up new paths for developing faster and more accurate algorithms for 3D reconstruction of scenes from images, building 3D maps, motion computation for autonomous driving, and in special effects for the movie industry.
The paper is co-authored by Timothy Duff (Georgia Tech), Kathén Kohn (KTH), and Anton Leykin (Georgia Tech) and was supported by ICERM (NSF DMS-1439786 and the Simons Foundation grant 507536) and IMPACT (reg. no. CZ.02.1.01/0.0/0.0/15 003/0000468) grant.
The ICCV conference (IEEE/CVF International Conference on Computer Vision) is a premier international computer vision conference and one of the top three conferences in the field (together with CVPR and ECCV). It is listed among the top 75 most cited journals and conferences over all areas of science by Google Scholar. A total of 1075 papers have been accepted this year from 4303 submissions (25% acceptance rate). The best papers including the Best Student Paper Award were selected by the Award Committee as the top 4 papers from all submissions. The best papers were presented to all 7500 attendees at a prize session during the opening of ICCV 2019.
[…] Tomas Pajdla from Applied Algebra and Geometry Group of the CIIRC CTU ranks as the 3rd top computer vision researcher in the Czech Republic according to Guide2Research. At the previous event, ICCV 2019, Tomáš Pajdla was one of the authors of the Best Student Paper. […]
Comments are closed.