Project Types

Updated on 17 Aug 2023
1 Minute to read

Print
Dark
Light
PDF

Article Summary

Share feedback

Thanks for sharing your feedback!

LandingLens offers these Project Types to help you label your images:

Object Detection
Classification
Segmentation
Visual Prompting

Select a Project Type When You Create a Project

Object Detection

Use to identify one or more objects in an image. Object Detection can be used to identify one or more objects within an image. Object Detection trains based on the labeled pixels (pixels inside the bounding box).

For more information, go to Object Detection.

Classification

Use to categorize (or "classify") the content of an image. Classification identifies the image as a whole. Classification trains based on all pixels in an image.

For more information, go to Classification.

Segmentation

Use to specify exact pixels to identify one or more regions within an image.

For more information, go to Segmentation.

Visual Prompting

Use to identify objects or areas in an image. You only need to label a few small areas for the Model to detect the whole object or area.

For more information, go to Visual Prompting.

Caution:

Visual Prompting is in Beta. It might not function as well as our other production-ready features.

Anomaly Detection

Use to identify deviations from the norm in an image.

For more information, go to Anomaly Detection.

Note:

Anomaly Detection is only available to legacy "classic" workflow users.

When to Use Each Project Type

Each Project Type is also designed for different use cases:

Project Type	When to Use	Examples
Object Detection	You want to identify multiple objects in an image. You want to identify one object in an image.	Identify multiple objects in an image. Scratches and missing parts on a laptop Apples and bananas Identify one object in an image. Deer Person License plate frame
Classification	You want to identify all the content within an image. You want to distinguish one object type from another.	Identify all content within an image. Image of a city Image of a basketball Distinguish one object type from another. Screws vs. nails Cats vs. dogs
Segmentation	You need to be precise. You want to identify one or more regions in an image.	Identify cracks on computer monitors
Visual Prompting	You want to create a Model quickly. You want to identify all regions in an image.	Identify objects with distinct textures. Animals Sand vs. rocks on a beach Identify objects that have irregular shapes. Regions of forests in satellite imagery Scratches on a product
Anomaly Detection	You have two sets of images: normal and abnormal. You want to teach the Model to identify abnormal objects without having to identify in advance all the possible abnormalities. Note: Anomaly Detection is only available to legacy "classic" workflow users.	Good food vs. rotten or spoiled food Good phone vs. broken phone

Was this article helpful?

What's Next

Object Detection