Show simple item record

dc.contributor.authorAlcazar, Juan Leon
dc.contributor.authorBravo, Maria A.
dc.contributor.authorThabet, Ali Kassem
dc.contributor.authorJeanneret, Guillaume
dc.contributor.authorBrox, Thomas
dc.contributor.authorArbelaez, Pablo
dc.contributor.authorGhanem, Bernard
dc.date.accessioned2019-12-18T11:00:01Z
dc.date.available2019-12-18T11:00:01Z
dc.date.issued2019-04-11
dc.identifier.urihttp://hdl.handle.net/10754/660665
dc.description.abstractInstance-level video segmentation requires a solid integration of spatial and temporal information. However, current methods rely mostly on domain-specific information (online learning) to produce accurate instance-level segmentations. We propose a novel approach that relies exclusively on the integration of generic spatio-temporal attention cues. Our strategy, named Multi-Attention Instance Network (MAIN), overcomes challenging segmentation scenarios over arbitrary videos without modelling sequence- or instance-specific knowledge. We design MAIN to segment multiple instances in a single forward pass, and optimize it with a novel loss function that favors class agnostic predictions and assigns instance-specific penalties. We achieve state-of-the-art performance on the challenging Youtube-VOS dataset and benchmark, improving the unseen Jaccard and F-Metric by 6.8% and 12.7% respectively, while operating at real-time (30.3 FPS).
dc.description.sponsorshipThis work was partially supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research, and by the German-Colombian Academic Cooperation between the German Research Foundation (DFG grant BR 3815/9-1) and Universidad de los Andes, Colombia.
dc.publisherarXiv
dc.relation.urlhttps://arxiv.org/pdf/1904.05847
dc.rightsArchived with thanks to arXiv
dc.titleMAIN: Multi-Attention Instance Network for Video Segmentation
dc.typePreprint
dc.contributor.departmentVisual Computing Center (VCC)
dc.contributor.departmentElectrical Engineering Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.eprint.versionPre-print
dc.contributor.institutionUniversidad de los Andes
dc.contributor.institutionUniversity of Freiburg
dc.identifier.arxivid1904.05847
kaust.personThabet, Ali Kassem
kaust.personGhanem, Bernard
refterms.dateFOA2019-12-18T11:00:59Z
kaust.acknowledged.supportUnitOffice of Sponsored Research


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
5.792Mb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record