4646088
doi
10.5281/zenodo.4646088
oai:zenodo.org:4646088
Asaf Peer
The Jackson Laboratory
Hao He
The Jackson Laboratory
Gautam Sabnis
The Jackson Laboratory
Vivek Kumar
The Jackson Laboratory
Mouse Grooming Detection Annotated Dataset
Brian Q Geuther
The Jackson Laboratory
doi:10.7554/eLife.63207
info:eu-repo/semantics/openAccess
Creative Commons Attribution Non Commercial No Derivatives 4.0 International
https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode
mouse
grooming
<p>This dataset contains 2 records. The first record is the annotated dataset. The second record contains a built singularity image containing the code and trained model for predicting on new videos.</p>
<p>We generated 1,253 video clips which total 2,637,363 frames. Each video had variable duration, depending upon the grooming prediction length. Annotators were required to provide a "Grooming" or "Not Grooming" annotation for each frame.<br>
The annotated dataset is stored in the h5 record and is described as follows:</p>
<ul>
<li>First level grouping is Train/Validation split</li>
<li>Second level grouping is by Video Clip</li>
<li>Each video contains 5 datasets
<ul>
<li>nframe
<ul>
<li>Number of frames in this video</li>
<li>Shape: 1</li>
</ul>
</li>
<li>video
<ul>
<li>Raw Video</li>
<li>Shape: nframes x 112 x 112</li>
</ul>
</li>
<li>label
<ul>
<li>Labels for each frame
<ul>
<li>0 = not grooming, 1 = grooming</li>
</ul>
</li>
<li>Shape: nframes</li>
</ul>
</li>
<li>mask
<ul>
<li>Information for whether or not annotators agreed
<ul>
<li>0 = disagree, 1 = agree</li>
<li>When annotators disagree, label contains the values from the first person to annotate the frame</li>
</ul>
</li>
<li>Shape: nframes</li>
</ul>
</li>
<li>nlabelers
<ul>
<li>Number of annotators that have labeled the video clip</li>
<li>Shape: 1 </li>
</ul>
</li>
</ul>
</li>
</ul>
Example usage of the trained model:
singularity run –nv GroomingInferRelease.sif Input_Movie.avi
The input movie must be a 480×480 video and appear visually similar to our training dataset. Since we do not employ distortion augmentation in this model, even slight differences in lighting can cause issues with performance.
Zenodo
2021-03-17
info:eu-repo/semantics/other
4646087
1617712040.98952
2615279616
md5:0b97bf5940fd9b64060c7187e40bca83
https://zenodo.org/records/4646088/files/GroomingInferRelease.sif
33090854078
md5:8b9a6b2132a99188ff9461d588d3e1d9
https://zenodo.org/records/4646088/files/GroomingDataset_2017-08-21.hdf5
public
10.7554/eLife.63207
Is cited by
doi
10.5281/zenodo.4646087
isVersionOf
doi