Multimedia Analysis

Hi! You've probably arrived here from the link on one of our papers. Here you will find a short collection of multimedia data related projects and publications. Click on one of the bookmarks for fast navigation to the publications page where you will find links to all information related to the paper.

Semantic Video Understanding
[CVPR2015] [CVPR2014] [ICMR2014]

Face and Person Identification, Retrieval
[ICMR2015] [FG2015] [AVSS2014] [CVPR2013] [CVPR2012]

or browse below to know about the status of code and data release.



Code and Data set downloads

Semantic Video Understanding

CVPR 2015
Alignment between novels and their TV series or film adaptations
, CVPR 2014
StoryGraphs: Codes to generate the visualization by loading all necessary data now available on Github (currently supports BBT, BUFFY)
ICMR 2014
Plot synopsis: alignment annotations on 22 episodes of BUFFY

Person Identification

ICMR 2015
Face tracks on all eight movies of the Harry Potter series to study effect of aging on id. Features (9GB+) can also be downloaded.
FG 2015
Improved weak labeling for person identification in TV series. Code available on Github
AVSS 2014
Large scale id: face tracks for season 5 all 22 episodes of BUFFY. Download single MAT file containing tracks for all episodes
CVPR 2013
Semi-supervised learning: face tracks on 6 episodes of BBT and BUFFY.
, CVPR 2012
PersonID: Preliminary helper code, face tracks, person tracks, video events on 6 episodes of BBT

Face Tracking

CVPR 2013
Download manually annotated face track bounding boxes for 6 episodes each of BBT and BUFFY which can be used for face tracking evaluation.

Links

At a glimpse

A snapshot of all multimedia related projects can be found here. This is the link you will come to from any of our papers too!

Multimedia Resources

I hope to collect links at one place to make it easy to browse the wonderful works.

CV, MM Papers on the Web

Meta-resources for Computer Vision conferences CVPapers, ACM Multimedia MM papers, and Multimedia Information Retrieval papers MIR.