Why Data Export Matters
Before you dive headfirst into the world of data annotation, it's crucial to consider two pivotal aspects:
With CVAT.ai, you have a plethora of export formats to choose from. Why is this essential? Because the format in which you export your data impacts its compatibility with machine learning models and frameworks. To give you a lay of the land, we offer an in-depth exploration of popular formats by task and use case in the video.
Know Before You Annotate
Before even starting your annotation journey, always consult the documentation. It provides an exhaustive list of export formats supported by CVAT.ai. This will help you decide which annotation types—be it bounding boxes, masks, polygons, or others—are best suited for your particular project. Different annotation types generate different dataset output options.
Putting Your Annotations to Work
Once you’ve chosen the appropriate annotation types and completed the annotation process, you can proceed to export your dataset. Here’s where the fun begins. You can export just one type of annotation or mixed annotations into a single file. If your project demands attribute-based labeling, you can also configure and seamlessly export these.
Configuring Attributes for Specific Formats
Our video tutorial includes a guide on how to export attributes and how they will appear in the exported files.
Video Annotation Exports
Video annotations come with their unique set of challenges and rewards. The tutorial covers how to export annotated videos in formats that either support or don’t support tracks. The choice of the format has significant implications on how the data can be utilized later.
Sounds interesting, right? Watch the full video tutorial here:
Not a CVAT.ai user? Click through and sign up here
Share your opinion and stay tuned!
Do not want to miss updates and news? Have any questions? Join our community: