Hello trend,
This is Satya Mallick from LearnOpenCV.com.
We are sharing a blog post on understanding Object Detection using CenterNet. CenterNet is an Anchor-free object detection model. Instead of predicting bounding boxes from anchor boxes, it predicts center points and regresses the box size to predict the bounding boxes.
In today's blog post, you will learn the following -
- What is object detection in general?
- Difference between anchor-based v/s anchor-free object detection?
- Advantages of anchor-free object detection.
- What is the CenterNet: Objects as Points algorithm, and how it works?
- How to use pre-trained models from the TensorFlow model zoo?
Without further ado, let's jump into the post
| Anchor-less Object Detection using CenterNet |
If you like our resources, do star us on GitHub to say thanks!
| Download Code (GitHub) |
You may also read our other posts on Anchor-free Object Detection:
- YOLOv6 Object Detection Paper Explanation and Inference
- YOLOX Object Detection Paper Explanation and Inference
P.S. There is another version of CenterNet that was released almost at the same time as the CenterNet model we discussed in today's post. It is named CenterNet: Keypoint Triplets for Object Detection.
Cheers!
Satya
Courses / YouTube / Facebook / LinkedIn / Twitter / Instagram
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.