EfficientDet outputs classes and bounding boxes. My question is about both but specifically I am interested in the class prediction net part. In the paper's diagram it shows 2 conv layers. I don't understand the code and how it works. And what's the difference between the 2 conv layers of classification and box prediction?
Faheem Ullah
Basically, The EfficientDet study basically employs "intuition" (and likely many, many development sets) to alter the topology of the NAS-FPN to arrive at the BiFPN, a bidirectional feature pyramid network. These BiFPN blocks are stacked on top of each other in the EfficientDet model. In the model scaling technique, the number of blocks changes.
Dear Faheem Ullah ,
Look the link, maybe useful.
https://ai.stackexchange.com/questions/33904/how-does-the-classification-head-of-efficientdet-work
Regards,
Shafagat
these articles explain the content precisely
I hope it will help you ,
https://ai.googleblog.com/2020/04/efficientdet-towards-scalable-and.html?m=1
https://paperswithcode.com/method/bifpn
I sent you link of some articles that explain the subject that you question about it
I hope it will help you ..
https://towardsdatascience.com/applied-deep-learning-part-4-convolutional-neural-networks-584bc134c1e2?gi=15b1095fb160
https://insightsimaging.springeropen.com/articles/10.1007/s13244-018-0639-9
- Badges
- Science topic