The DynTex database Homepage
The DynTex Database
The DynTex database is a diverse collection of high-quality dynamic
texture videos. Currently we
are finalizing the structure of the database. In the meantime, more than 650
sequences are available for downloading.
For an overview of the total contents of the DynTex database, we provide a page
of key frames of all sequences currently available. To be able to play
the corresponding sample video files the password obtained at
registration is required.
Central to the database is the so-called golden set of high-quality dynamic texture sequences that satisfy all criteria of the acquisition protocol. This set will be further structured by its underlying physical processes, e.g turbulent motion, waving motion, discrete units etc. Additionally, they will be classified by their shot type, i.e. either as closeup (shot consisting entirely of the dynamic texture, no segmentation required), or as context (the dynamic texture is shown in its context). Next to the golden set we will offer various other sets. Examples are a collection of dynamic texture sequences recorded with a moving camera, and a collection with several dynamic textures per sequence.
Central to the database is the so-called golden set of high-quality dynamic texture sequences that satisfy all criteria of the acquisition protocol. This set will be further structured by its underlying physical processes, e.g turbulent motion, waving motion, discrete units etc. Additionally, they will be classified by their shot type, i.e. either as closeup (shot consisting entirely of the dynamic texture, no segmentation required), or as context (the dynamic texture is shown in its context). Next to the golden set we will offer various other sets. Examples are a collection of dynamic texture sequences recorded with a moving camera, and a collection with several dynamic textures per sequence.
Classification Benchmark
In order to use DynTex as a benchmark for recognition,
different sub-datasets have been compiled and labeled. Only a single
dynamic texture is present in each sequence with no panning or zooming
performed by the camera (see the Classification benchmark page).Sequence Annotation
Next to the basic
structuring of the
database, an extensive annotation effort has been performed. For each
sequence
of the golden set,
we offer
an XML-description characterizing both its content, as
well as the context in which it was recorded.
Examples of annotation variables are: presence of disturbance, spatial and temporal regularity, texture scale. A description of the annotation process can be found at the annotation page.
The XML-annotations can be used for convenient selection of interesting subsets of the DynTex database.
For more information, please refer to:
DynTex: A Comprehensive Database of Dynamic Textures.
Renaud Péteri, Sándor Fazekas and Mark J. Huiskes.
Pattern Recognition Letters, Volume 31, Issue 12, 1 September 2010, Pages 1627-1632.
Examples of annotation variables are: presence of disturbance, spatial and temporal regularity, texture scale. A description of the annotation process can be found at the annotation page.
The XML-annotations can be used for convenient selection of interesting subsets of the DynTex database.
For more information, please refer to:
DynTex: A Comprehensive Database of Dynamic Textures.
Renaud Péteri, Sándor Fazekas and Mark J. Huiskes.
Pattern Recognition Letters, Volume 31, Issue 12, 1 September 2010, Pages 1627-1632.
Sequence Format
The sequences are available in avi-files (type II) using lossless
DV-compression. A free Panasonic DV codec can be downloaded
here.
The sequences are in color, PAL format (720 x 576, 25
fps). Next to these 'raw' sequences, processed sequences, which have
been deinterlaced
with a
spatio-temporal median filter, are also provided. Please refer to the Download
section for further details.
All sequences consist of at least 250 frames. If sequences of a
standard length are required, the first 250 frames of each sequence can
be used. Sound is included only when relevant to the texture.
Previews
Below are ten preview samples from the DynTex database.
By clicking a key frame, a DivX-compressed video of the
sequence can be viewed. Note that, due to high compression, the
quality of these videos is much lower than the official DynTex
sequences. For downloading of the full quality videos registration is
required. Please visit the Download
section for further instructions.
Remarks?
Questions? Broken links? 
