EfficientDet is an efficient object detection model that achieves a very high mAP at a fraction of the compute requirements of other object detection models. It is made with a similar philosophy to that of EfficientNet with some additional methods for better multi-scale feature fusion. This is a follow up video of 'MobileNetV2 and EfficientNet' where we explored EfficientNet’s scaling methodology and the MBConv blocks it is made of in detail.
MobileNetV2 and EfficientNet Video: https://youtu.be/IBndcd4UfTs
EfficientDet: Scalable and Efficient Object Detection:https://arxiv.org/abs/1911.09070
Credits: I would like to thank Deepak Anand for his discussions on this topic. Some slides have been in part or entirely taken from Jinwon Lee’s presentation of EfficientDet on SlideShare. Thanks to Jinwon for open sourcing his presentation.
Chapters
0:00 Introduction
1:04 Challenges and Related Works
4:13 Model Scaling and MBConv Recap
6:50 Architecture overview
7:46 BiFPN
12:14 Weight Feature Fusion
16:18 EfficientDet Scaling
19:36 Results
