Published December 7, 2021 | Version v1
Dataset Open

data SLEAP

  • 1. University of Konstanz, Germany
  • 2. Tel Aviv University


The data includes a 
1) raw video of the timelapse for each analysis: 

2) slp_Files: this folder contains all the.slp files for each video analysis. These are the files that can be loaded into SLEAP and contain the 5,10 or 20 labelled training frames. From these files users can train the models either directly in the GUI or on Google CoLab.

3) AnalysisFiles_UserLabels: This folder contains the output analysis file with only the user labelled training frames. Here we outputted directly all 20 frames. These can be also directly generated from the .slp files in the slp_Files folder

4) AnalysisFiles_PredictedLables: This folder contains the output analysis files with the predicted data for each video analysis. Note that we used a cleaning step before conducting out analyses, as IDs might be named differently in the predicted compared to the user labelled analysis files.


Files (92.5 MB)

Name Size Download all
92.5 MB Preview Download