SSD: Single Shot MultiBox Detector
Wei Liu1 Dragomir Anguelov2 Dumitru Erhan3 Christian Szegedy3 Scott Reed4 Cheng-Yang Fu1 Alexander C. Berg1
Abstract
We present a method for detecting objects in images using a single deep neural network. Our approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location. At prediction time, the network generates scores for the presence of each object category in each default box and produces adjustments to the box to better match the object shape. Additionally, the network combines predictions from multiple feature maps with different resolutions to naturally handle objects of various sizes. SSD is simple relative to methods that require object proposals because it completely eliminates proposal generation and subsequent pixel or feature resampling stages and encapsulates all computation in a single network. This makes SSD easy to train and straightforward to integrate into systems that require a detection component. Experimental results on the PASCAL VOC, COCO, and ILSVRC datasets confirm that SSD has competitive accuracy to methods that utilize an additional object proposal step and is much faster, while providing a unified framework for both training and inference. For $300\times 300$ input, SSD
中文速览
目标检测领域长期被"先生成候选框、再逐框分类"的两阶段流程主导，速度慢、难以实时应用。SSD（Single Shot MultiBox Detector）提出用一个卷积神经网络直接完成检测，在不同分辨率的特征图上预先铺设多种尺度和长宽比的默认框（default boxes），网络只需一次前向传播就能同时输出每个框的类别得分和位置偏移，彻底省去候选区域生成和特征重采样两个耗时步骤。实验结果显示，在 PASCAL VOC 2007 测试集上，300×300 输入的 SSD 以 59 FPS 的速度达到 74.3% mAP，既比同为单阶段的 YOLO 精度高出约 11 个百分点，速度又比两阶段的 Faster R-CNN 快近 8 倍。这项工作的意义在于首次证明单阶段检测器可以在精度上与两阶段方法比肩，为实时、高精度目标检测提供了一个简洁统一的框架，对后续检测器设计影响深远。
原文 arXiv:1512.02325;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1512.02325v5