Anestis Papazoglou, Luca Del Pero, Vittorio Ferrari

University of Edinburgh (CALVIN)

Overview

This dataset contains video shots for two different classes: tigers and cars. We collected the shots from 188 car ads (~1–2 min each) and 14 nature documentaries about tigers (~40 min), amounting to roughly 14 h of video. We automatically partitioned these raw videos into shorter shots, and kept only those showing at least one instance of the class. This produced 806 shots for the car and 1880 for the tiger class, typically 1–100 sec in length.

We annotated aspect labels as follows. First, we randomly chose five frames per shot, and annotated each of them with the number of objects shown. We then gave aspect labels only to frames showing exactly one object (to avoid ambiguities). This produces a total of 6610 frames with aspect label for tigers, and 3485 for cars.

We also provide foreground segmentation masks computed using the software by Papazoglou and Ferrari. For more details, see the README file.

Aspect annotations

Part visibility

Head viewpoint (for tigers)		Leg action (for tigers)

Downloads: Version 1.0

Filename	Description	Release Date	Size
README.txt	Description of contents	11 September 2016	7.7 KB
cars.tar.gz	Car videos and annotations	11 September 2016	1.7 GB
tigers.tar.gz	Tiger videos and annotations	11 September 2016	44.9 GB
segmentations_cars.tar.gz	Segmentations for the car videos	11 September 2016	43.7 MB
segmentations_tigers.tar.gz	Segmentations for the tiger videos	11 September 2016	254 MB

Citations

We release this dataset together with our Image and Vision Computing 16 paper on automatic aspect discovery from video. If you use this dataset for your research, please cite:

@JOURNAL{papazoglou16imavis,
author = {Papazoglou, A. and Del Pero, L. and Ferrari, V.},
title = {Discovering object aspects from video},
journal = {Image and Vision Computing},
year = {2016}
}

Important Notice

These videos were downloaded from the internet, and may subject to copyright. We don’t own the copyright of the videos and only provide them for non-commercial research purposes.