"task_id": "MFcoPiJDQ7Ti1mKlSRbe2K6AhalHyvR5",
    "type": undefined,
    "tag": "Draw lines on the train tracks",
    "data": {
      attachment: "https://dummy.com/image.png",
      annotation_categories: ["track"]
    "importance": "standard",
    "status": "pending",
    "response": {
      "annotation": {
        "size": {
          "depth": "3"
          "width": "800"
          "height": "534"
        "object": [
            "name": undefined,
            "pose": "Unspecified",
            "line": {
              "x1": "509",
              "x2": "466",
              "y1": "467",
              "y2": "440"
            "occuluded": "0",
            "difficuilt": "0",
            "truncated": "0"
        "filename": "image.png",
        "segmented": "0",
    "completed_at": null,
    "credit_cost": null
Leverage our human workers to delegate line annotation tasks using our extremely extensive and simple API.
Use Cases
Self Driving Cars
With the ability to support any categories, Raidon's API provides the perfect platform to annotate images that may be used to power self driving cars and other AI automation needs such as planes, boats and robots. With support for the PASCAL VOC format our annotations can easily integrate into Google's Tensforflow.

Motion Tracking
Line annotation can be used to power machine learning solutions to track humans potentially providing applications in the use of games, retail and security. Motion tracking can play a vital rule in recognising changes in position and then react accordingly to the movement of humans or objects.

Boundary Tracking
Track whether a particular object is within a fixed boundary. This can be used in sports such as football to check whether a ball remains within the field and potentially be used to call offside by comparing the position of defenders of one team and the strikers of another team.

Why Raidon?
With free repetition allowance, ability to download VOC complaint XML files and superb customer service we're confident you'll love us.
Pixel Perfect Annotation
Whether you want tight or loose lines, our trained workers will be able to meet your requirements. With our simple and easy to understand API, you'll be provided with every vertex of the line.
PASCAL VOC Annotations
All of our annotations are PASCAL VOC complaint allowing you to easily download an XML file that you can feed directly to leading machine-learning frameworks such as Tensorflow for object detection.
Cost Free Repetition
If you don't agree with the annotation of one of our workers or we didn't meet your requirement just contact support and we'll repeat your annotations free of charge. We're not happy unless you're happy.
Supports all files
Whether you want to provide us a url to image you want to annotate or upload a file to our servers, Raidon supports all file types. As long as we can open the file we'll annotate it.
Edit your tasks
Accidently submitted the wrong image or file? No problem - Raidon provided free cost edits allowing you to change the type of annotation, file submitted or a tasks importance allowing you to speed its annotation up!
See Annotations in Dashboard
Forget the terminal, with Raidon's Dashboard you'll be able to view the lines drawn for each image annotated, send images back for reannotation, download results in JSON or XML format and much more!
How it Works
Just send us your unannotated images and get back VOC complaint annotated lines
   curl -X POST \
      https://raidon.io/api/v1/task/annotation/ \
      -H 'Authorization: Token 49fa77ade8a3b9504a8d5138836030b44dd9ea46' \
      -d type='lines' \
      -d tag='Tag the markings' \
      -d attachment='https://dummyimageurl.com/drones.jpg' \
      -d categories='marking'
Execute Code
   import requests 
   headers = {
      'Authorization': 'Token 49fa77ade8a3b9504a8d5138836030b44dd9ea46'
   data = {
      'type': 'lines', 
      'tag': 'Tag the markings', 
      'attachment': 'https://dummyimageurl.com/drones.jpg', 
      'categories': ['marking]
   r = requests.post('https://raidon.io/api/v1/task/annotation/',
Execute Code
   const request = await fetch('https://raidon.io/api/v1/task/annotation', method={
   method: 'POST', 
   headers: {
         'Authorization': 'Token 49fa77ade8a3b9504a8d5138836030b44dd9ea46', 
   body: {
         type: 'lines',
         tag: 'Tag the markings',
         attachment: 'https://dummyimageurl.com/drones.jpg', 
         categories: ['marking']
   const response = await request.json()
Execute Code
      "task_id": "2rXrvjvYST8dQppTFHA9Ni68Tz3X8DpM",
      "created_at": "2018-08-24T20:59:05.374168Z",
      "type": "lines",
      "tag": "Tag the markings",
      "data": {
         "attachment": "http://media-raidon.s3-us-west-2.amazonaws.com/2rXrvjvYST8dQppTFHA9Ni68Tz3X8DpM/drones.jpg",
         "categories": [
      "importance": "standard",
      "status": "completed",
      "callback_url": "www.raidon.io/api/v1/task/2rXrvjvYST8dQppTFHA9Ni68Tz3X8DpM",
      "response": {
         "annotation": {
               "size": {
                  "depth": "3",
                  "width": "1800",
                  "height": "1200"
               "object": [
                  "line": {
                     "x1": "671",
                     "x2": "672",
                     "y1": "446",
                     "y2": "452"
                  "name": "marking",
                  "pose": "Unspecified",
                  "occluded": "0",
                  "difficult": "0",
                  "truncated": "0"
            "filename": "drones.jpg",
            "segmented": "0"
      "completed_at": "2018-08-24T21:02:46.146796Z",
      "credit_cost": "$0.28"
Raidon charges a fixed price for each image and each line annotated so you know exactly what you're paying for. There’s no monthly charge, no inactivity charges and absolutely no hiddent costs. It's just simple and transparent pricing.
$0.08 + $0.05
per annotation
Frequently Asked Questions
What is Raidon?

Raidon provides machine learning firms with the ability to leverage our human workers to annotate images and other tasks to power their AI products speeding up their route to market.

What are lines?

Lines refer to the shape of each annotation on the image described by a group of x and y cordinates. Each x cordinate respresents the pixel distance between the line vertex and the left border of the image and each y cordinate represents the pixel distance between the line vertex and the top border of the image. More information can be found in our API docs.

Do I get a free trial?

Yes, new users are given the ability to request the annotation of 20 images to be annotated by lines, without being charged or their bank details being asked for, to try out our services. Early stage machine learning startup can request to have their free limit increased to 100 by contacting sales.

What is an urgency task?

Urgency tasks are given priority over other tasks and we aim to complete urgency tasks within 1 working day. However, urgency tasks are subject to different pricing compared to standard tasks

How long do standard tasks take?

Standard task will vary depending on the on the number of images sent to be annotated as well as the number of categories, however, we aim to to complete standard tasks within 3 working days.

How do I send the images to be annotated?

You can send images either as a url or upload the image from your local storage to our API. In either case your images will be saved on our servers.

Do you sell our images to third parties?

Absolutely not, Raidon has a strict policy not to sell our clients data to third parties. We only store your uploaded data for your use so you can confirm the annotations using our dashboard. All images are deleted if the task is deleted via an API request. Furthermore, all data associated with the user is deleted if the user makes a requests to have his account terminated indefinitely.

How does the pricing work?

Raidon charges $0.08 for each image sent to be annoated and a further $0.10 for each annotation. An annotation is defined as actual blue boxes drawed on top of the images you see above.

How do I view the annotations once they're completed?

You can either view each task you sent to our API to be annotated, using our interative dashboard or get the annotation back using our API either as a JSON response or a VOC complaint XML file.

What other annotation types do you support?

Plans to support the COCO and KITTI format are currently being worked on. Sign up and create an account to recieve updates.