当前位置: X-MOL 学术Appl. Intell. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Fast multi-resolution occlusion: a method for explaining and understanding deep neural networks
Applied Intelligence ( IF 5.3 ) Pub Date : 2020-11-04 , DOI: 10.1007/s10489-020-01946-3
Hamed Behzadi-Khormouji , Habib Rostami

Deep Convolutional Neural Networks (DCNNs) contain a high level of complexity and nonlinearity, so it is not clear based on what features DCNN models make decisions and how they can reach such promising results. There are two types of visualization techniques to interpret and explain the deep models: Backpropagation-based and Perturbation-based algorithms. The most notable drawback of the backpropagation-based visualization is that they cannot be applied for all architectures, whereas Perturbation-based visualizations are totally independent of the architectures. These methods, however, take a lot of computation and memory resources which make them slow and expensive, thereby unsuitable for many real-world applications. To cope with these problems, in this paper, a perturbation-based visualization method called Fast Multi-resolution Occlusion (FMO) are presented which is efficient in terms of time and resource consumption and can be considered in real-world applications. In order to compare the FMO with five well-known Perturbation-based visualizations methods such as Occlusion Test, Super-pixel perturbation (LIME), Randomized Input Sampling (RISE), Meaningful Perturbation and Extremal Perturbation, different experiments are designed in terms of time-consumption, visualization quality and localization accuracy. All methods are applied on 5 well-known DCNNs DenseNet121, InceptionV3, InceptionResnetV2, MobileNet and ResNet50 using common benchmark datasets ImageNet, PASCAL VOC07 and COCO14. According to the experimental results, FMO is averagely 2.32 times faster than LIME on five models DenseNet121, InceptionResnetV2, InceptionV3, MobileNet and ResNet50 with images of ILSVRC2012 dataset as well as 24.84 times faster than Occlusion Test, 11.87 times faster than RISE, 8.72 times faster than Meaningful Perturbation and 10.03 times faster than Extremal Perturbation on all of the five used models with images of common dataset ImageNet without scarifying visualization quality. Moreover, the methods are evaluated in terms of localization accuracy on two hard common datasets of PASCAL VOC07 and COCO14. The results show that FMO outperforms the compared relevant methods in terms of localization accuracy. Also, FMO extends the superimposing process of the Occlusion Test method, which yields a heatmap with more visualization quality than the Occlusion Test on many colorful images.



中文翻译:

快速多分辨率遮挡:一种解释和理解深度神经网络的方法

深度卷积神经网络(DCNN)包含高度的复杂性和非线性,因此,基于DCNN模型的决策功能以及如何获得如此有希望的结果,尚不清楚。有两种类型的可视化技术可以解释和解释深层模型:基于反向传播和基于摄动的算法。基于反向传播的可视化的最显着缺点是不能将它们应用于所有体系结构,而基于扰动的可视化完全独立于体系结构。但是,这些方法占用大量计算和内存资源,这使其运行缓慢且昂贵,因此不适用于许多实际应用。为了解决这些问题,在本文中,提出了一种基于扰动的可视化方法,称为快速多分辨率遮挡(FMO),该方法在时间和资源消耗方面均十分有效,并且可以在实际应用中考虑使用。为了将FMO与五种著名的基于扰动的可视化方法进行比较,例如遮挡测试,超像素扰动(LIME),随机输入采样(RISE),有意义的扰动和极值扰动,在时间方面设计了不同的实验-消费,可视化质量和本地化准确性。使用通用基准数据集ImageNet,PASCAL VOC07和COCO14,将所有方法应用于5个著名的DCNN DenseNet121,InceptionV3,InceptionResnetV2,MobileNet和ResNet50。根据实验结果,在五个型号DenseNet121上,FMO平均比LIME快2.32倍,在所有五个使用图像的模型上,具有ILSVRC2012数据集的图像的InceptionResnetV2,InceptionV3,MobileNet和ResNet50以及图像比Occlusion Test快24.84倍,比RISE快11.87倍,比Meaningful Perturbation快8.72倍,并且比Extremal Perturbation快10.03倍。通用数据集ImageNet,而不会影响可视化质量。此外,在PASCAL VOC07和COCO14的两个硬通用数据集上,根据定位精度评估了这些方法。结果表明,就定位精度而言,FMO优于已比较的相关方法。此外,FMO扩展了遮挡测试方法的叠加过程,该方法在许多彩色图像上生成的热图比“遮挡测试”具有更高的可视化质量。在所有五个使用通用数据集ImageNet的图像模型上,具有ILSVRC2012数据集图像的MobileNet和ResNet50分别比遮挡测试快24.84倍,比RISE快11.87倍,比有意义摄动快8.72倍和比极值摄动快10.03倍而不会影响可视化质量。此外,在PASCAL VOC07和COCO14的两个硬通用数据集上,根据定位精度评估了这些方法。结果表明,就定位精度而言,FMO优于已比较的相关方法。此外,FMO扩展了遮挡测试方法的叠加过程,该方法在许多彩色图像上生成的热图比“遮挡测试”具有更高的可视化质量。在所有五个使用通用数据集ImageNet的图像模型上,具有ILSVRC2012数据集图像的MobileNet和ResNet50分别比遮挡测试快24.84倍,比RISE快11.87倍,比有意义摄动快8.72倍和比极值摄动快10.03倍而不会影响可视化质量。此外,在PASCAL VOC07和COCO14的两个硬通用数据集上,根据定位精度评估了这些方法。结果表明,就定位精度而言,FMO优于已比较的相关方法。此外,FMO扩展了遮挡测试方法的叠加过程,该方法在许多彩色图像上生成的热图比“遮挡测试”具有更高的可视化质量。在所有五个使用通用数据集ImageNet的图像的模型上,均比“有意义摄动”快72倍,比“极端摄动”快10.03倍,而不会影响可视化质量。此外,在PASCAL VOC07和COCO14的两个硬通用数据集上,根据定位精度评估了这些方法。结果表明,就定位精度而言,FMO优于已比较的相关方法。此外,FMO扩展了遮挡测试方法的叠加过程,该方法在许多彩色图像上生成的热图比“遮挡测试”具有更高的可视化质量。在所有五个使用通用数据集ImageNet的图像的模型上,均比“有意义摄动”快72倍,比“极端摄动”快10.03倍,而不会影响可视化质量。此外,在PASCAL VOC07和COCO14的两个硬通用数据集上,根据定位精度评估了这些方法。结果表明,就定位精度而言,FMO优于已比较的相关方法。此外,FMO扩展了遮挡测试方法的叠加过程,该方法在许多彩色图像上生成的热图比“遮挡测试”具有更高的可视化质量。在PASCAL VOC07和COCO14的两个硬通用数据集上,根据定位精度评估了这些方法。结果表明,就定位精度而言,FMO优于已比较的相关方法。此外,FMO扩展了遮挡测试方法的叠加过程,该方法在许多彩色图像上生成的热图比“遮挡测试”具有更高的可视化质量。在PASCAL VOC07和COCO14的两个硬通用数据集上,根据定位精度评估了这些方法。结果表明,就定位精度而言,FMO优于已比较的相关方法。此外,FMO扩展了遮挡测试方法的叠加过程,该方法在许多彩色图像上生成的热图比“遮挡测试”具有更高的可视化质量。

更新日期:2020-11-05
down
wechat
bug