How to train the model to detecte objects only?

[JingangLang]

How can I close the branch used to Object Segmentation task?I only have images and the corresponding border and category information.Hence how to modify the codes to only detecte the objects?Thanks!

[vcarpani]

If you only want to detect objects, without segmentation, you can use our implementation of retinanet: https://github.com/fizyr/keras-retinanet

[JingangLang]

First of all, thank you for your timely reply.I have tried retinanet to detecte the car and the shop logos. Among them, the vehicle detection effect is very good, but the brand detection is not satisfactory. After debugging the parameters, the test score reaches the bottleneck area, so we are looking for a new scheme. I am looking forward to your valuable advice. Thank you!

[hgaiser]

Let's move this discussion to the keras-retinanet slack channel.

[JingangLang]

Okay.