Consistent Optimization for Single-Shot Object Detection

Abstract

We present consistent optimization for single stage object detection. Previous works of single stage object detectors usually rely on the regular, dense sampled anchors to generate hypothesis for the optimization of the model. Through an examination of the behavior of the detector, we observe that the misalignment between the optimization target and inference configurations has hindered the performance improvement. We propose to bride this gap by consistent optimization, which is an extension of the traditional single stage detector's optimization strategy. Consistent optimization focuses on matching the training hypotheses and the inference quality by utilizing of the refined anchors during training. To evaluate its effectiveness, we conduct various design choices based on the state-of-the-art RetinaNet detector. We demonstrate it is the consistent optimization, not the architecture design, that yields the performance boosts. Consistent optimization is nearly cost-free, and achieves stable performance gains independent of the model capacities or input scales. Specifically, utilizing consistent optimization improves RetinaNet from 39.1 AP to 40.1 AP on COCO dataset without any bells or whistles, which surpasses the accuracy of all existing state-of-the-art one-stage detectors when adopting ResNet-101 as backbone. The code will be made available.

Technical report

References in corpus (11)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• YOLOv3: An Incremental Improvement
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• DSSD : Deconvolutional Single Shot Detector
• Cascade R-CNN: Delving into High Quality Object Detection
• Beyond Skip Connections: Top-Down Modulation for Object Detection
• Deformable ConvNets v2: More Deformable, Better Results
• MetaAnchor: Learning to Detect Objects with Customized Anchors
• Single-Shot Bidirectional Pyramid Networks for High-Quality Object Detection

Cited by in corpus (10)

• RepPoints: Point Set Representation for Object Detection
• ROAD: The ROad event Awareness Dataset for Autonomous Driving
• Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection
• RefineDetLite: A Lightweight One-stage Object Detection Framework for CPU-only Devices
• Dynamic Anchor Learning for Arbitrary-Oriented Object Detection
• Improving Multispectral Pedestrian Detection by Addressing Modality Imbalance Problems
• CenterFace: Joint Face Detection and Alignment Using Face as Point
• ACFD: Asymmetric Cartoon Face Detector
• Semi-Anchored Detector for One-Stage Object Detection
• STELA: A Real-Time Scene Text Detector with Learned Anchor