Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Detailed instruction #159

Open
pnambiar opened this issue Dec 28, 2018 · 2 comments
Open

Detailed instruction #159

pnambiar opened this issue Dec 28, 2018 · 2 comments

Comments

@pnambiar
Copy link

I am new to this tool. I find the instructions a bit confusing. All I want to do is annotate a list of sequential images ( Not using mechanical turk)

  1. Do I have to convert it to mp4. Is there an option to use the list of images directly?
  2. I have put the videos in the the static folder.
    Then what? Could you please provide step by step instructions? That will be really helpful.
@antingshen
Copy link
Owner

  1. You can use list of images directly.
  2. Put the images in the static folder. Create your video object that lists the paths (URL) to the images. Then follow the other instructions for usage.

@clayshieh
Copy link
Collaborator

The instructions could be a bit clearer on non mechanical turk use cases but here are the step by step instructions if you're using a mac:

In terminal:

  1. Clone this repository with git clone https://github.com/antingshen/BeaverDam.git
  2. cd BeaverDam
  3. Make sure Python 3 is installed.
    If not: brew install python3 (Mac) or sudo apt-get install python3 (Ubuntu)
  4. Make sure virtualenv is installed.
    If not: pip3 install virtualenv or maybe sudo pip3 install virtualenv
  5. If you're using a mac then edit requirements.txt and replace all lines with the following:
Django==1.10
sqlparse==0.2.2
tqdm==4.8.4
uWSGI==2.0.17
markdown==2.6.7
  1. Run scripts/setup to setup virtualenv and install all dependencies
  2. Run scripts/seed -f to download sample seed data
  3. Run source venv/bin/activate to activate virtualenv
  4. Run scripts/serve to run the application
  5. Navigate to localhost:5000 in your browser to see what the application looks like (creds are test/password)

The seed data uses videos hosted on s3 but it also downloads and places a local file named 0.mp4 in the folder (/annotator/static/videos/) which contain the videos you want to use.

Adding this seeded video is a good exercise of how you can add your own videos locally.
Navigate to localhost:5000/admin in your browser (login again with test/password if necessary)
Under the Annotator table click Videos (note the +Add button for later, its a shortcut).

You should now be at http://localhost:5000/admin/annotator/video/
This page shows you all the videos available for annotating. You can click the link in the ID column to view the details and you can click the link the VIDEO column to bring you to the actual page to annotate the video. To add the 0.mp4 video we talked about before, click the Add Video + button in the top right of the page which is equivalent to that shortcut we talked about earlier.

The fields of interest are Filename (the actual filename of the video/image) which in this case would be 0.mp4 and Host (the hostname of where the file is located) which in this case would be /static/videos/ because we're running on locally. If you want additional labels for this particular video you can also add labels in the labels field.

Picture here:
screen shot 2019-01-14 at 6 32 09 pm

Click save.

You should now see the new entry that was created (/video/149/). Click on the link for video 149 under the VIDEO column which should now bring you to an annotator screen showing the 0.mp4 video.

If you see a grey screen with a loading screen of 3 dots then you might have done something wrong.

Hope these instructions help!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants