Title:
The Middle Child Problem: Revisiting Parametric Min-cut and Seeds for Object Proposals

Simple item page
dc.contributor.author Humayun, Ahmad
dc.contributor.author Li, Fuxin
dc.contributor.author Rehg, James M.
dc.contributor.corporatename Georgia Institute of Technology. Institute for Robotics and Intelligent Machines en_US
dc.contributor.corporatename Georgia Institute of Technology. College of Computing en_US
dc.contributor.corporatename Oregon State University en_US
dc.date.accessioned 2016-08-18T17:34:14Z
dc.date.available 2016-08-18T17:34:14Z
dc.date.issued 2015-12
dc.description © 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works. en_US
dc.description DOI: 10.1109/ICCV.2015.187 en_US
dc.description.abstract Object proposals have recently fueled the progress in detection performance. These proposals aim to provide category-agnostic localizations for all objects in an image. One way to generate proposals is to perform parametric min-cuts over seed locations. This paper demonstrates that standard parametric-cut models are ineffective in obtaining medium-sized objects, which we refer to as the middle child problem. We propose a new energy minimization framework incorporating geodesic distances between segments which solves this problem. In addition, we introduce a new superpixel merging algorithm which can generate a small set of seeds that reliably cover a large number of objects of all sizes. We call our method POISE - "Proposals for Objects from Improved Seeds and Energies." POISE enables parametric min-cuts to reach their full potential. On PASCAL VOC it generates ~2,640 segments with an average overlap of 0.81, whereas the closest competing methods require more than 4,200 proposals to reach the same accuracy. We show detailed quantitative comparisons against 5 state-of-the-art methods on PASCAL VOC and Microsoft COCO segmentation challenges. en_US
dc.identifier.citation Humayun, A., Li, F., & Rehg, J. M. (2015). The Middle Child Problem: Revisiting Parametric Min-Cut and Seeds for Object Proposals. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 2015, pp. 1600-1608. en_US
dc.identifier.doi 10.1109/ICCV.2015.187 en_US
dc.identifier.uri http://hdl.handle.net/1853/55478
dc.language.iso en_US en_US
dc.publisher Georgia Institute of Technology en_US
dc.publisher.original Institute of Electrical and Electronics Engineers en_US
dc.subject Category-agnostic localizations en_US
dc.subject Geodesic distance en_US
dc.subject Middle child problem en_US
dc.subject Object proposals en_US
dc.subject Parametric min-cut model en_US
dc.subject POISE en_US
dc.subject Proposals for objects from improved seeds and energies en_US
dc.title The Middle Child Problem: Revisiting Parametric Min-cut and Seeds for Object Proposals en_US
dc.type Text
dc.type.genre Proceedings
dspace.entity.type Publication
local.contributor.author Rehg, James M.
local.contributor.corporatename Institute for Robotics and Intelligent Machines (IRIM)
local.contributor.corporatename College of Computing
relation.isAuthorOfPublication af5b46ec-ffe2-4ce4-8722-1373c9b74a37
relation.isOrgUnitOfPublication 66259949-abfd-45c2-9dcc-5a6f2c013bcf
relation.isOrgUnitOfPublication c8892b3c-8db6-4b7b-a33a-1b67f7db2021
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
The Middle Child Problem- Revisiting Parametric Min-cut and Seeds for Object Proposals.pdf
Size:
5.2 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
3.13 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Research Publications