EfficientDet: Scalable and Efficient Object Detection

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 multiscale 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 better 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 55.1 AP on COCO test-dev with 77M parameters and 410B FLOPs, being 4x - 9x smaller and using 13x - 42x fewer FLOPs than previous detectors. Code is available at https://github.com/google/automl/tree/master/efficientdet.

CVPR 2020

Cited by in corpus (28)

• A Simple Semi-Supervised Learning Framework for Object Detection
• Hierarchical Neural Architecture Search for Deep Stereo Matching
• DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
• DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis
• Affinity and Diversity: Quantifying Mechanisms of Data Augmentation
• DFUC2020: Analysis Towards Diabetic Foot Ulcer Detection
• Empirical Upper Bound in Object Detection and More
• Knowing what you know: valid and validated confidence sets in multiclass and multilabel prediction
• Auto-Panoptic: Cooperative Multi-Component Architecture Search for Panoptic Segmentation
• ASFD: Automatic and Scalable Face Detector
• Tails: Chasing Comets with the Zwicky Transient Facility and Deep Learning
• Scene Graph Reasoning for Visual Question Answering
• LeYOLO, New Embedded Architecture for Object Detection
• Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search
• Spectro-Temporal RF Identification using Deep Learning
• FNAS: Uncertainty-Aware Fast Neural Architecture Search
• DriftNet: Aggressive Driving Behavior Classification using 3D EfficientNet Architecture
• Cheaper Pre-training Lunch: An Efficient Paradigm for Object Detection
• Effective Fusion Factor in FPN for Tiny Object Detection
• A Feasible Framework for Arbitrary-Shaped Scene Text Recognition
• Pretrained Encoders are All You Need
• Multilevel Knowledge Transfer for Cross-Domain Object Detection
• PatchNet -- Short-range Template Matching for Efficient Video Processing
• On the Visual-based Safe Landing of UAVs in Populated Areas: a Crucial Aspect for Urban Deployment
• Unsupervised data augmentation for object detection
• Proactive Network Maintenance using Fast, Accurate Anomaly Localization and Classification on 1-D Data Series
• DICE: Deep Significance Clustering for Outcome-Aware Stratification
• Video based real-time positional tracker