Multi-Scale Attention Deep Neural Network for Fast Accurate Object Detection
Multi-Scale Attention Deep Neural Network for Fast Accurate Object Detection
复制标题
用于快速准确目标检测的多尺度注意力深度神经网络
DOI：
10.1109/tcsvt.2018.2875449
复制
发表时间：
2019-10-01
影响因子：
8.4
通讯作者：
Yin, Zhouping
中科院分区：
文献类型：
--
作者：
Song, Kaiyou;Yang, Hua;Yin, Zhouping
由于聚类背景、遮挡、截断和尺度变化导致物体外观发生巨大变化，因此在计算机视觉中，目标检测仍然是一项具有挑战性的任务。目前基于深度神经网络（DNN）的目标检测方法无法同时实现高精度和高效率。为了克服这一限制，本文提出了一种新的多尺度注意深度神经网络，用于高精度的目标检测。本文提出的MSA-DNN方法利用一种新型的多尺度特征融合模块（MSFFM）构建高级语义特征。在此基础上，提出了一种基于融合层的MSA模块（MSAM），利用图像级标签的全局语义信息指导检测。一方面，mffm可以捕获全局语义信息，进一步增强由MSFFM构建的融合层的语义特征表示，从而提高检测精度。另一方面，利用MSAM生成的MSA地图可以快速、粗略地定位不同尺度的目标。此外，引入了基于注意力的硬负性挖掘策略，过滤掉负样本，减少了搜索空间，极大地缓解了严重的类不平衡问题。在具有挑战性的PASCAL VOC 2007， PASCAL VOC 2012和MS COCO数据集上的广泛实验结果表明，MSA-DNN在保持高效率的同时实现了最先进的检测精度。此外，MSA-DNN显著提高了小目标检测精度。
Object detection remains a challenging task in computer vision due to the tremendous extent of changes in the appearances of objects caused by clustered backgrounds, occlusion, truncation, and scale change. Current deep neural network (DNN)-based object detection methods cannot simultaneously achieve a high accuracy and a high efficiency. To overcome this limitation, in this paper, we propose a novel multi-scale attention (MSA) DNN for accurate object detection with high efficiency. The proposed MSA-DNN method utilizes a novel multi-scale feature fusion module (MSFFM) to construct high-level semantic features. Subsequently, a novel MSA module (MSAM) based on the fused layers of the MSFFM is introduced to exploit the global semantic information of image-level labels to guide detection. On the one hand, MSAM can capture global semantic information to further enhance the semantic feature representation of the fused layers constructed by the MSFFM, thereby improving the detection accuracy. On the other hand, the MSA maps generated by MSAM can be employed to rapidly and coarsely locate objects at different scales. In addition, an attention-based hard negative mining strategy is introduced to filter out negative samples to reduce the search space, dramatically alleviating the severe class imbalance problem. Extensive experimental results on the challenging PASCAL VOC 2007, PASCAL VOC 2012, and MS COCO datasets demonstrate that MSA-DNN achieves a state-of-the-art detection accuracy while maintaining a high efficiency. Furthermore, MSA-DNN significantly improves the small-object detection accuracy.