Non-parametric contextual relationship learning for semantic video object segmentation
Tutkimustuotos › › vertaisarvioitu
Yksityiskohdat
Alkuperäiskieli | Englanti |
---|---|
Otsikko | Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - 23rd Iberoamerican Congress, CIARP 2018, Proceedings |
Toimittajat | Ruben Vera-Rodriguez, Julian Fierrez, Aythami Morales |
Kustantaja | Springer Verlag |
Sivut | 325-333 |
Sivumäärä | 9 |
ISBN (painettu) | 9783030134686 |
DOI - pysyväislinkit | |
Tila | Julkaistu - 2019 |
OKM-julkaisutyyppi | A4 Artikkeli konferenssijulkaisussa |
Tapahtuma | Iberoamerican Congress on Pattern Recognition - Madrid, Espanja Kesto: 19 marraskuuta 2018 → 22 marraskuuta 2018 |
Julkaisusarja
Nimi | Lecture Notes in Computer Science |
---|---|
Vuosikerta | 11401 |
ISSN (painettu) | 0302-9743 |
ISSN (elektroninen) | 1611-3349 |
Conference
Conference | Iberoamerican Congress on Pattern Recognition |
---|---|
Maa | Espanja |
Kaupunki | Madrid |
Ajanjakso | 19/11/18 → 22/11/18 |
Tiivistelmä
We propose a novel approach for modeling semantic contextual relationships in videos. This graph-based model enables the learning and propagation of higher-level spatial-temporal contexts to facilitate the semantic labeling of local regions. We introduce an exemplar-based nonparametric view of contextual cues, where the inherent relationships implied by object hypotheses are encoded on a similarity graph of regions. Contextual relationships learning and propagation are performed to estimate the pairwise contexts between all pairs of unlabeled local regions. Our algorithm integrates the learned contexts into a Conditional Random Field (CRF) in the form of pairwise potentials and infers the per-region semantic labels. We evaluate our approach on the challenging YouTube-Objects dataset which shows that the proposed contextual relationship model outperforms the state-of-the-art methods.