A Better Detection Head | Essentials of Object Detection
HTML-код
- Опубликовано: 9 сен 2024
- This is a continuation of the Detection Head tutorial that explains how to write the code such that you can avoid ugly indexing into the tensors and also have more maintainable and extensible components.
It would beneficial to first watch the DetectionHead tutorial
Link to the DetectionHead tutorial:
• Detection Head | Essen...
Link to the Google Colab notebook:
colab.research...
How can we use these (predictions_box, predictions_obj, predictions_cls) information from the decoder and create visualizations on input image ?
I think in this case, it's faster to cluster everything in a single conv layer but I agree separating the modules is better practices
Correct
Hello, I still didnt understand how to we combine the three detection layers output to get one final result of tensor! or is there any other technique ?
You reshape them as (batch, num_anchors, predictions) and then concatenate them.
@@KapilSachdeva thank you