Introduction 
Annotating pictures for object detection or occasion segmentation generally is a tedious and costly course of. Nevertheless, latest developments in AI have paved the best way for faster and extra correct annotation strategies. Grounding DINO and Section Something Mannequin (SAM) are two state-of-the-art fashions that may significantly velocity up this course of. On this complete weblog submit, we’ll display how these fashions may be utilized for picture annotation and the conversion of object detection datasets into occasion segmentation datasets.
Now we have ready a Jupyter Pocket book that you would be able to open domestically or in Google Colab and observe alongside whereas studying the weblog submit.
Let’s dive into the main points!
Annotating Photographs with Grounding DINO and SAM
Grounding DINO and SAM are highly effective AI fashions that may help within the dataset annotation course of. Grounding DINO is able to zero-shot detection of any object within the picture, whereas SAM can convert these bounding bins into occasion segmentation masks.
