English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Simple Does It: Weakly Supervised Instance and Semantic Segmentation

Khoreva, A., Benenson, R., Hosang, J., Hein, M., & Schiele, B. (2016). Simple Does It: Weakly Supervised Instance and Semantic Segmentation. Retrieved from http://arxiv.org/abs/1603.07485.

Item is

Basic

show hide
Genre: Paper
Other : Weakly Supervised Semantic Labelling and Instance Segmentation

Files

show Files
hide Files
:
1603.07485v2 (Preprint), 7MB
Name:
1603.07485v2
Description:
-
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-

Locators

show

Creators

show
hide
 Creators:
Khoreva, Anna1, Author           
Benenson, Rodrigo1, Author           
Hosang, Jan1, Author           
Hein, Matthias2, Author
Schiele, Bernt1, Author           
Affiliations:
1Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society, ou_1116547              
2External Organizations, ou_persistent22              

Content

show
hide
Free keywords: Computer Science, Computer Vision and Pattern Recognition, cs.CV
 Abstract: Semantic labelling and instance segmentation are two tasks that require particularly costly annotations. Starting from weak supervision in the form of bounding box detection annotations, we propose to recursively train a convnet such that outputs are improved after each iteration. We explore which aspects affect the recursive training, and which is the most suitable box-guided segmentation to use as initialisation. Our results improve significantly over previously reported ones, even when using rectangles as rough initialisation. Overall, our weak supervision approach reaches ~95% of the quality of the fully supervised model, both for semantic labelling and instance segmentation.

Details

show
hide
Language(s): eng - English
 Dates: 2016-03-242016-11-232016-03-24
 Publication Status: Published online
 Pages: 21 p.
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: arXiv: 1603.07485
URI: http://arxiv.org/abs/1603.07485
BibTex Citekey: Khoreva1603.07485
 Degree: -

Event

show

Legal Case

show

Project information

show

Source

show