Welcome to the support page for the Pose Data Annotator app. You can find the most frequent questions here.
If you have a question, feel free to contact us at [email protected] and [email protected]. If you have a suggestion, feature request, or want to file a bug report, proceed here.
- Open the app.
- Select the Annotate video section in the left sidebar.
- Drop a file into the box, or select one using the Load Video button.
- Tap Start Processing and wait for the results. A popup will open once the process is done and you will be able to save the resulting
csv
file.
- Open the app.
- Select the Annotate dataset section in the left sidebar.
- Drop a folder into the box, or select one using the Load Dataset button.
- Tap Start Processing and wait for the results. A popup will open once the process is done and you will be able to save the resulting
csv
file.
The app utilizes the Vision engine for body, hand and face pose estimation. The videos are split into frames individually annotated and outputted for any potential experiments or use-cases involving pose data.
Open App Preferences (by pressing Cmd + ,
). Here you can customize the FPS to be used, which parts of the body to annotate, and more.