LabelMe video: Building a video database with human annotations

Author

Yuen, Jenny ; Russell, Bryan ; Liu, Ce ; Torralba, Antonio

Author_Institution

MIT, Cambridge, MA, USA

fYear

2009

fDate

Sept. 29 2009-Oct. 2 2009

Firstpage

1451

Lastpage

1458

Abstract

Currently, video analysis algorithms suffer from lack of information regarding the objects present, their interactions, as well as from missing comprehensive annotated video databases for benchmarking. We designed an online and openly accessible video annotation system that allows anyone with a browser and internet access to efficiently annotate object category, shape, motion, and activity information in real-world videos. The annotations are also complemented with knowledge from static image databases to infer occlusion and depth information. Using this system, we have built a scalable video database composed of diverse video samples and paired with human-guided annotations. We complement this paper demonstrating potential uses of this database by studying motion statistics as well as cause-effect motion relationships between objects.

Keywords

Internet; image motion analysis; image segmentation; object recognition; video databases; video signal processing; Internet; cause-effect motion relationships; diverse video samples; human-guided annotations; image databases; labelme video; motion statistics; real-world videos; video analysis algorithms; video annotation system; video database; Algorithm design and analysis; Computer vision; Humans; Image databases; Internet; Layout; Shape; Statistics; Tracking; Videoconference;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Vision, 2009 IEEE 12th International Conference on

Conference_Location

Kyoto

ISSN

1550-5499

Print_ISBN

978-1-4244-4420-5

Electronic_ISBN

1550-5499

Type

conf

DOI

10.1109/ICCV.2009.5459289

Filename

5459289