How to: Design a Video Object Tracking Job

Data

Data preparation for a Video Object Tracking job is simple but has a few key aspects outlined below to ensure the data is processed correctly and is able to be annotated. In order to get started here is what you’ll need:

Video

The video files should:

1. Be hosted and publicly viewable
   • If there are any security concerns your Customer Success Manager can help configure a secure storage solution
   • In MP4 or AVI format
2. Broken into sensible rows of work for a contributor to complete
   • We recommend as a best practice to have about 100 frames per video. Depending on your frame rate this will be between 3 and 10 second clips per video.

The video files should not:

1. Exceed 100mb in size or 1800 frames

A CSV:

1. At least one column with the link of the video to be annotated with a column header (Ex. “video-url”)
   • As needed, you can pass any other metadata along as columns

Instructions

1. Make an instructional video
   • Creating a video will help you understand the tool and discover some of the edge cases in your data. It will also give contributors context on how the videos should be annotated.
2. Provide guidance on how the tool works
   • The video annotation tool has built in features to help contributors annotate more efficiently. This includes a full menu of hotkeys and tooltips. Feel free to copy paste these tips into your instructions:

CML

CML for a Video Object Tracking job is available to our Enterprise Figure Eight users. Please contact your Customer Success Manager for access to this code. The product is in BETA, so please consider the following:

1. The job needs to be designed in CML and there is currently no graphical editor for this tool
2. Launching this job requires one of our trusted video annotation channels. Please reach out to your Customer Success Manager to set this up.
3. If you need any help, don’t hesitate to reach out to help@figure-eight.com.

Parameters

Below are the parameters available for the cml:video_shapes tag. Some are required in the element, some can be left out.

• label-required="true"
  • You will need to create or upload an ontology of at least one class
• label-required="false"
  
  • There’s no ontology in the left sidebar, you don’t need to create one, and users just create boxes with no labels
• name: the name of your output column
• video-url: the column header containing the video to be annotated in the source file
• assistant="linear_interpolation" or "object_tracking":
  • There are two different types of machine assistance to create annotations:
    • Object Tracking is ideal when:
      • The camera is moving
      • Note: This is only available for bounding boxes. If any other shapes are being used, your job must use Linear Interpolation. 
    • Linear Interpolation is ideal when:
      • The objects are moving in a linear fashion
      • The camera is stationary
      • The objects being tracked are small and often change in size
        
• validates-required: contributors must view all frames of a video
  • Note: Preview of the job and tool will not work before launch. The frames need preprocessing before they can be loaded and this processing will not begin until the job is launched.
• type: accepts a comma-separated array of any of the four shapes types. 
  • Example: type="['box','polygon','dot','line']

Settings

• Necessary settings:
  • 1 row per page
  • 1 judgment per row
  • At least 3 hours per assignment
    • This can be set via the API using the following command or by contacting help@figure-eight.com
      • Set Time Per Assignment
      • curl -X PUT --data-urlencode "job[options][req_ttl_in_seconds]={n}" "https://api.figure-eight.com/v1/jobs/{job_id}.json?key={api_key}"

Some considerations prior to Launch

Unlike a typical Figure Eight job, once the job is launched we will pre-process the video data linked in the job. While this is occurring for each row the row will be in state “preprocessing” before becoming “judgable”. 

If the unit cannot be preprocessed it will be automatically canceled. This is to prevent contributors from seeing a broken tool and collecting annotations on incorrectly formatted data. Some common reasons a video row may be canceled are:

• The video file is too large or contains too many frames
• The URL provided does not lead to a visible video file - either the permissions are incorrect or the file is otherwise corrupted

If this occurs and you’re able to identify and correct the issue you can re-upload the video and order a judgment on the new rows.

Monitor Quality

• In the BETA version of this tool, monitoring is limited to reviewing finished videos and their results
• Peer review is a way to quality check annotations done through the Video Bounding Box tool. Please contact your Customer Success Manager to enable this review as needed.
  • To set up a peer review job you will need:
    • The original video link in a column, and a column which is the result of your first job
  • Important note: Please be sure to use the same ontology as in the original job for peer review. You can download it from the original job and upload it to the peer review job when needed, or make a copy of the original job with no data.