EfficientDet: Scalable and Efficient Object Detection
Citations Over TimeTop 1% of 2020 papers
Abstract
Model efficiency has become increasingly important in computer vision. In this paper, we systematically study neural network architecture design choices for object detection and propose several key optimizations to improve efficiency. First, we propose a weighted bi-directional feature pyramid network (BiFPN), which allows easy and fast multi-scale feature fusion; Second, we propose a compound scaling method that uniformly scales the resolution, depth, and width for all backbone, feature network, and box/class prediction networks at the same time. Based on these optimizations and EfficientNet backbones, we have developed a new family of object detectors, called EfficientDet, which consistently achieve much better efficiency than prior art across a wide spectrum of resource constraints. In particular, with single-model and single-scale, our EfficientDet-D7 achieves state-of-the-art 52.2 AP on COCO test-dev with 52M parameters and 325B FLOPs1, being 4x - 9x smaller and using 13x - 42x fewer FLOPs than previous detector. Code is available at https://github.com/google/ automl/tree/master/efficientdet.
Related Papers
- → TFPN: Twin Feature Pyramid Networks for Object Detection(2019)21 cited
- → Intelligent Detection of Missing and Unattended Objects in Complex Scene of Surveillance Videos(2012)10 cited
- → Object detection for rotated and densely arranged objects in aerial images using path aggregated feature pyramid networks(2020)3 cited
- → Feature detectors by autoencoders: Decomposition of input patterns into atomic features by neural networks(1995)4 cited
- → ExtraDet With Quad Feature Pyramid Networks(2020)1 cited