from DeepLabCut

To import a DeepLabCut dataset, follow the steps below. Note that you can only label one folder at a time.

1. In the menu, click on: Import > Load DeepLabCut dataset
2. Navigate to your DeepLabCut project
3. Go within the labeled-data
4. Select a folder and click “Upload”

If there are no labels yet, you may need to specify bodyparts (below). Note: Due to limitations on h5 libraries in javascript, just the h5 (not images) is sent to server to convert it to a csv to parse.

from Lightning Pose

To import a Lightning Pose dataset, follow the steps below.

1. In the menu, click on: Import > Load Lightning Pose dataset
2. Navigate to your folder which contains “CollectedData.csv”
3. Select that folder and click “Upload”

from SLEAP

To import a SLEAP dataset, follow these steps:

1. In the menu, click on: Import > Load SLEAP dataset (SLP file)
2. Select your slp file and select “Upload”

Note that SLP file must have the images padded with zeros to account for the javascript h5 loading library. This is already in the latest SLEAP code and will be available in future SLEAP releases. For now, you can use this python script to pad your video items.

annotating images

Place keypoints by clicking in the image. The keypoints will be placed following the order of bodyparts. (See above for how to edit the bodyparts.)

You can move a keypoint by clicking it and dragging it. You can delete a keypoint by selecting it and either pressing “d” or clicking the X icon in the toolbar.

To navigate across images, you can use the left/right arrow keys, clicking in the file menu, or click the left/right arrows in the toolbar.

labeling multiple animals

Anivia supports labeling multiple animals, but only if you export to a SLEAP file currently.

The interface focuses on one animal “instance” at a time. By default when you click to place points it places them in instance 0.

Here is how you navigate the instance interface: - “i” for new instance (or circle plus in toolbar) - “u” to delete an instance (or circle minus in toolbar) - up and down arrows to switch between instances (or circle up and down in toolbar, or just click on a point from that instance)

to DeepLabCut

To export your annotations to DeepLabCut, follow these steps:

1. In the menu, click on: Export > Export Annotations (DeepLabCut format)
2. Save your file as “CollectedData_SCORER.h5” within the folder within labeled-data that you imported Replace SCORER above with the actual scorer name you specified at the top of your DeepLabCut config.yaml

Note: Similarly to importing, due to limitations on h5 libraries in javascript, the labels are sent to server to convert them to generate an h5 file. Again, no images are ever sent out of your machine.

to Lightning Pose

To export your annotations to Lightning Pose, follow these steps:

1. In the menu, click on: Export > Export Annotations (Lightning Pose format)
2. Save your file as “CollectedData.csv” within the folder you imported

to SLEAP

To export your annotations to Lightning Pose, follow these steps:

1. In the menu, click on: Export > Export Annotations (SLP format)
2. Save your slp file wherever.

It might take a bit of time (less than a minute) to generate the SLP file for export.