mldr.datasets (version 0.3.1)

mediamill: Dataset with features extracted from video sequences and semantic concepts assigned as labels

Description

Multilabel dataset from the video domain.

Usage

mediamill()

format

An mldr object with 43907 instances, 120 attributes and 101 labels

source

Snoek, C. G. M. and Worring, M. and van Gemert, J. C. and Geusebroek, J. M. and Smeulders, A. W. M., "The challenge problem for automated detection of 101 semantic concepts in multimedia", in Proc. 14th ACM International Conference on Multimedia, MULTIMEDIA06, pp. 421-430, 2006

Examples

mediamill()  # Check and load the dataset
toBibtex(mediamill)
mediamill$measures