In this paper we consider the problem of motion segmentation, where only pairwise correspondences are assumed as input without prior knowledge about tracks. The problem is formulated as a two-step process. First, motion segmentation is performed on image pairs independently. Secondly, we combine independent pairwise segmentation results in a robust way into the final globally consistent segmentation. Our approach is inspired by the success of averaging methods. We demonstrate in simulated as well as in real experiments that our method is very effective in reducing the errors in the pairwise motion segmentation and can cope with large number of mismatches.
Robust motion segmentation from pairwise matches / Arrigoni, Federica; Pajdla, Tomas. - (2019), pp. 671-681. (Intervento presentato al convegno 17th IEEE/CVF International Conference on Computer Vision, ICCV 2019 tenutosi a Seoul nel 27th October - 2nd November 2019) [10.1109/ICCV.2019.00076].
Robust motion segmentation from pairwise matches
Arrigoni Federica;
2019-01-01
Abstract
In this paper we consider the problem of motion segmentation, where only pairwise correspondences are assumed as input without prior knowledge about tracks. The problem is formulated as a two-step process. First, motion segmentation is performed on image pairs independently. Secondly, we combine independent pairwise segmentation results in a robust way into the final globally consistent segmentation. Our approach is inspired by the success of averaging methods. We demonstrate in simulated as well as in real experiments that our method is very effective in reducing the errors in the pairwise motion segmentation and can cope with large number of mismatches.File | Dimensione | Formato | |
---|---|---|---|
ICCV19.pdf
Open Access dal 01/01/2022
Tipologia:
Post-print referato (Refereed author’s manuscript)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
6.65 MB
Formato
Adobe PDF
|
6.65 MB | Adobe PDF | Visualizza/Apri |
09008798.pdf
Solo gestori archivio
Tipologia:
Versione editoriale (Publisher’s layout)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
500.45 kB
Formato
Adobe PDF
|
500.45 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione