DL之DSSD:DSSD算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略
相关文章
DL之DSSD:DSSD算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略
DL之DSSD:DSSD算法的架构详解
DSSD算法的简介(论文介绍)
DSSD,是在SSD上做的改进,即Deconvolutional Single Shot Detector,反卷积单步骤探测器。
Abstract
The main contribution of this paper is an approach for introducing additional context into state-of-the-art general object detection. To achieve this we first combine a state-ofthe-art classifier (Residual-101 [14]) with a fast detection framework (SSD [18]). We then augment SSD+Residual101 with deconvolution layers to introduce additional largescale context in object detection and improve accuracy, especially for small objects, calling our resulting system DSSD for deconvolutional single shot detector. While these two contributions are easily described at a high-level, a naive implementation does not succeed. Instead we show that carefully adding additional stages of learned transformations, specifically a module for feed-forward connections in deconvolution and a new output module, enables this new approach and forms a potential way forward for further detection research. Results are shown on both PASCAL VOC and COCO detection. Our DSSD with 513 × 513 input achieves 81.5% mAP on VOC2007 test, 80.0% mAP on VOC2012 test, and 33.2% mAP on COCO, outperforming a state-of-the-art method R-FCN [3] on each dataset.
本文的主要贡献是将附加上下文引入到最先进的一般对象检测中。为了实现这一点,我们首先结合了一个最先进的分类器(Residual-101[14])和一个快速检测框架(SSD[18])。然后,我们使用反褶积层来增加SSD+Residual101,以在目标检测中引入额外的大范围上下文,并提高精度,特别是对于小对象,调用我们得到的系统DSSD来实现反卷积单镜头检测器。虽然这两个贡献很容易在高层进行描述,但是一个简单的实现是不会成功的。相反,我们展示了详细添加额外的学习转换阶段,特别是反褶积中的前馈连接模块和一个新的输出模块,使这种新方法成为可能,并为进一步的检测研究形成了一个潜在的前进方向。结果表明,PASCAL VOC和COCO 检测。我们的513×513输入的DSSD在VOC2007测试中实现了81.5%的mAP,在VOC2012测试中实现了80.0%的mAP,在COCO上实现了33.2%的mAP,在每个数据集上都优于目前最先进的R-FCN[3]方法。
Conclusion
We propose an approach for adding context to a stateof-the-art object detection framework, and demonstrate its effectiveness on benchmark datasets. While we expect many improvements in finding more efficient and effective ways to combine the features from the encoder and decoder, our model still achieves state-of-the-art detection results on PASCAL VOC and COCO. Our new DSSD model is able to outperform the previous SSD framework, especially on small object or context specific objects, while still preserving comparable speed to other detectors. While we only apply our encoder-decoder hourglass model to the SSD framework, this approach can be applied to other detection methods, such as the R-CNN series methods [12, 11, 24], as well.
我们提出了一种将上下文添加到最先进的对象检测框架的方法,并在基准数据集上证明了它的有效性。虽然我们期望在寻找更有效和更有效的方法来结合编码器和解码器的特性方面有许多改进,但我们的模型仍然在PASCAL VOC和COCO上实现了最先进的检测结果。我们的新DSSD模型能够超越以前的SSD框架,特别是在小对象或特定上下文对象上,同时仍然保持与其他检测器相当的速度。虽然我们只将我们的编解码器沙漏模型应用于SSD框架,但是这种方法也可以应用于其他检测方法,比如R-CNN系列方法[12,11,24]。
论文
Cheng-Yang Fu , Wei Liu , Ananth Ranga, AmbrishTyagi , Alexander C. Berg .
DSSD : Deconvolutional Single Shot Detector,CVPR 2017
https://arxiv.org/abs/1701.06659
1、DSSD框架结构
残差网络上的SSD和DSSD网络:蓝色模块是SSD框架中添加的层,称之为SSD层。在下图中,红色图层是DSSD层。
2、DSSD模型的特点、贡献
- SSD算法对小目标不够鲁棒(会出现误检和漏检);最主要的原因是浅层特征图的表示能力不够强。DSSD算法的核心思想就是提高浅层的表示能力。
- DSSD在原来的SSD模型上主要作了两大改进:
一是替换掉VGG,而改用了Resnet-101作为特征提取网络并在对不同尺度的特征图进行默认框检测时使用了更新的检测单元;
二则在网络的后端使用了多个反卷积层(deconvolution layers)以有效地扩展低维度信息的上下文信息(contextual information) ,从而有效地改善了小尺度目标的检测。
DSSD算法的架构详解
更新……
DSSD算法的案例应用
更新……