RWTH-PHOENIX-Weather 2014: Continuous Sign Language Recognition Dataset


Human Language Technology & Pattern Recognition Group

RWTH Aachen University, Germany


In Short:

you can download here the RWTH-PHOENIX Weather 2014 (53GB) Continuous Sign Language Benchmark data set. If you use it in your research, please cite:

Detailed Description:

Over a period of three years (2009 - 2011) the daily news and weather forecast airings of the German public tv-station PHOENIX featuring sign language interpretation have been recorded. Currently, only the weather forecasts of a subset of 386 editions have been transcribed using gloss notation. The transcriptions have been carried out by deaf and hard-of-hearing native speakers of German sign language. Additionally, the spoken German weather forecast has been transcribed in a semi-automatic fashion using the RASR speech recognition system. Moreover, an additional translation of the glosses into spoken German has been created to capture allowable translation variability.

The signing is recorded by a stationary color camera placed in front of the sign language interpreters. Interpreters wear dark clothes in front of an artificial grey background with color transition. All recorded videos are at 25 frames per second and the size of the frames is 210 by 260 pixels. Each frame shows the interpreter box only.

Due to legal constraints RWTH cannot publish the original annotation files in the ELAN xml format and the recorded video sequences. Instead xml files containing ground-truth gloss annotation with corresponding id as well as the image sequences corresponding to these ids are provided.


RWTH-PHOENIX Weather 2014 Contents:
    1. README
      1. README
          1. File 1
      1. README


Published State of the Art Continuous Sign Language Recognition Results on RWTH-PHOENIX-Weather 2014 Multisigner (in Word Error Rate, the lower the better):
Author WER Dev[%] WER Test [%]
Koller, Zargaran and Ney, CVPR 2017 27.1 26.8
Huang, Zhou, Zhang, Li and Li, AAAI 2018 -- 38.3
Cui, Liu and Zhang, CVPR 2017 39.4 38.7
Camgoz, Hadfield, Koller and Bowden, ICCV 2017 40.8 40.7
Koller, Zargaran, Ney and Bowden, BMVC 2016 38.3 38.8
Koller, Ney and Bowden, CVPR 2016 47.1 45.1
Koller, Forster and Ney, CVIU 2015 57.3 55.6

Published Signer Independent Results on the RWTH-PHOENIX-Weather 2014 Signerindependent SI05 (in Word Error Rate, the lower the better):
Author WER Dev[%] WER Test [%]
Koller et al, CVPR 2017 45.1 44.1

Oscar Koller 2017-11-05
Creative Commons License