Scale 101: Task Types

Scale has over 20 annotation types to choose from, we've written this to help you choose the best one for your needs.

Both On-Demand and Enterprise customers have access to:

  • Scale Image: Comprehensive annotations for images
  • Scale Data Classification: Categorize a variety of media

Our Enterprise customers have access to:

  • Scale Video: Comprehensive annotations for videos
  • Scale 3D + Sensor Fusion: Advanced annotations for LiDAR + RADAR data
  • Scale Text: Sophisticated annotations for text-based data
  • Scale Document: Secure processing of documents

Enterprise Task Types

If you want to learn more about task types that are only available on our Enterprise customers, please contact our sales team.

Scale Image

Comprehensive annotations for images.

Image Annotation

In an imageannotation task, one of our Taskers will view the given image and be able to annotate

  • bounding boxes (optionally, rotated)
  • polygons
  • cuboids
  • ellipsis / ovals
  • lines
  • points

Semantic Segmentation Annotation

In a segmentannotation task, one of our Taskers will view the given image and classify every pixel of the image according to the labels provided. You will get a full semantic, pixel-wise, dense segmentation of the image.

Scale Data Classification

Categorize a variety of media

Categorization

In a categorization task, one of our Taskers will view the attachment and select the correct answer for every taxonomy/question. Example use cases are spam detection, copyright detection, product categorization, etc.

Attachments can be text, images, videos, websites, pdf files, or audio.

Scale Video

Enterprise-Only - Comprehensive annotations for videos. To learn more, please contact our sales team.

Video Annotation

In a videoannotation task, one of our Taskers will view the given video and be able to annotate

  • bounding boxes (optionally, rotated)
  • polygons
  • cuboids
  • ellipsis / ovals
  • lines
  • points

Do keep in mind each frame labeled is equivalent to one image.

Video Playback

In a videoplaybackannotation task, we will view the given video file and draw bounding boxes around the specified objects, returning the positions and sizes of these boxes.

In addition to annotating objects in the scene, you can also request to annotate specific events that happen during the video. The main difference between events and annotations is that events are not linked to a specific object or shape, but rather to a frame or specific timestamp.

Scale 3D + Sensor Fusion

Enterprise-Only - Advanced annotations for LiDAR + Radar + Camera data. To learn more, please contact our sales team.

Lidar Annotation

In a lidarannotation task, one of our Taskers views output from a series of LIDAR frames, along with optional radar and camera data, and annotates where different objects exist in the 3D space with 3D cuboids. To learn more, please contact our sales team.

Lidar Segmentation

In a lidarsegmentation task, one of our Taskers views output from a series of LIDAR frames, along with optional camera data, and annotates where different objects exist in the 3D space by assigning a class to each lidar point. To learn more, please contact our sales team.

2D / 3D Linking

In a lidarlinking task, one of our Taskers takes an already completed lidar annotation task and adjusts the projected cuboids in the camera images, creating a linking of 2D and 3D data. To learn more, please contact our sales team.

Scale Text

Enterprise-Only - Comprehensive annotations for documents. To learn more, please contact our sales team.

Named Entity Recognition

In a namedentityrecognition task, one of our Taskers will read the provided text and highlight any text entity mentions that correspond to the specified labels.

Text Collection

In a textcollection task, one of our Taskers will collect information from the given attachments and/or through the web following the instructions that you provide. Example use cases include labeling structurally-complex data from an attached image or querying sentiment information given a set of links.

Scale Document

Enterprise-Only - Comprehensive annotations for documents. To learn more, please contact our sales team.

Document Transcription

In a documenttranscription task, one of our Taskers will look at single or multiple page documents in order to transcribe and annotate any information you would like. Example use cases include transcribing invoices from PDF or manually scraping a screenshot for information.

Don't see a Task Type that fits your needs?

We'd love to brainstorm with you and hear about your use-case, please get in touch with us to discuss.

Updated 3 months ago