Blog

The latest from Google Research

EfficientDet: Towards Scalable and Efficient Object Detection

Wednesday, April 15, 2020

Posted by Mingxing Tan, Software Engineer and Adams Yu, Research Scientist, Google Researchroboticsdriverless carshigh-accuracy detectorsEfficientDet: Scalable and Efficient Object DetectionCVPR 2020EfficientNetEfficientDet

EfficientDet architecture. EfficientDet uses EfficientNet as the backbone network and a newly proposed BiFPN feature network.

Model Architecture Optimizationsbackbonefeature networkclass/box networkResNetsResNeXtAmoebaNetEfficientNetFPNPANetNAS-FPNFPNPANetNAS-FPN

A comparison between our BiFPN and previous feature networks. Our BiFPN allows features (from the low resolution P3 levels to high-resolution P7 levels) to repeatedly flow in both top-down and bottom-up ways.

depthwise separable convolutionsprevious workModel PerformanceCOCO datasetmean average precision

EfficientDet achieves state-of-the-art 52.2 mAP, up 1.5 points from the prior state of the art (not shown since it is at 3045B FLOPs) on COCO test-dev under the same setting. Under the same accuracy constraint, EfficientDet models are 4x-9x smaller and use 13x-42x less computation than previous detectors.

semantic segmentationPascal VOC 2012

EfficientDet achieves better quality on Pascal VOC 2012 val than DeepLabV3+ with 9.8x less computation, under the same setting without COCO pre-training.

Open SourceGitHubAcknowledgementsThanks to the paper co-authors Ruoming Pang and Quoc V. Le. We thank Daiyi Peng, Golnaz Ghiasi, Tianjian Meng for their help on infrastructure and discussion. We also thank Adam Kraft, Barret Zoph, Ekin D. Cubuk, Hongkun Yu, Jeff Dean, Pengchong Jin, Samy Bengio, Tsung-Yi Lin, Xianzhi Du, Xiaodan Song, and the Google Brain team.