Start your first Object Detection project on Roboflow.

Dataset Upload --> Dataset Generation & Model Training with Roboflow Train. Plus: Quick insight into Dataset Health Check, selecting preprocessing & augmentation steps, exporting in different annotation formats, and deploying your model to run inference.

Topics Covered:

Creating Your Project

• Workspaces, Projects, and Versions‍ 

What are annotation batches? Uploading images (working with annotation batches, and our annotation workflow)

Getting Images from Roboflow Universe to add to a project

• How to Use Roboflow Universe‍ | What is Roboflow Universe?

Uploading images to your annotation batch (fresh unannotated images, already annotated images, video files)

• Deleting Annotation Batches‍ 

Using the annotation tool

• The Labeling Interface‍ 
• Labeling Guide: Object Detection‍ 

Completing an annotation batch, and generating versions of your dataset for model training

Uploading the images from Roboflow Universe to the project

Navigating annotation batches, and uploading new images

• Deleting Annotation Batches‍ 

Generating versions: A quick primer on Preprocessing and Augmentation

• More on Preprocessing and Augmentation

Primer on Dataset Health Check

• Dataset insights, such as the total number of annotations, heatmap of annotations, average image sizes, and the number of labels for each class (class balance)
• Dataset Health Check: Improving Your Dataset‍
• Health Check: Filter by Class‍ 

Training your model

• Roboflow Train | Choosing Between Computer Vision Model Sizes

Using Label Assist

Training a custom model with the Roboflow Model Library

Roboflow Train results: model inference/deployment options and running tests

Next Steps:

Testing and Deploying Your Model

Deployment: Using the Roboflow Inference API‍ 

Python Package for OAK Deployment‍ 

Implementing Active Learning‍