{"title":"基于 CNN 的生成式对抗网络性能评估指标","authors":"Adarsh Prasad Behera;Satya Prakash;Siddhant Khanna;Shivangi Nigam;Shekhar Verma","doi":"10.1109/TAI.2024.3401650","DOIUrl":null,"url":null,"abstract":"In this work, we propose two convolutional neural network (CNN) based metrics, classification score (CS) and distribution score (DS), for performance evaluation of generative adversarial networks (GANs). Though GAN-generated images can be evaluated through manual assessment of visual fidelity, it is prolonged, subjective, challenging, tiresome, and can be misleading. Existing quantitative methods are biased toward memory GAN and fail to detect overfitting. CS and DS allow us to experimentally prove that training of GANs is actually guided by the dataset that it improves with every epoch and gets closer to following the distribution of the dataset. Both methods are based on GAN-generated image classification by CNN. CS is the root mean square (rms) value of three different classification techniques, direct classification (DC), indirect classification (IC), and blind classification (BC). It exhibits the degree to which GAN can learn the features and generate fake images similar to real datasets. DS shows the contrast between the mean distribution of GAN-generated data and the real data. It indicates the extent to which GANs can create synthetic images with similar distribution to real datasets. We evaluated CS and DS metrics for different variants of GANs and compared their performances with existing metrics. Results show that CS and DS can evaluate the different variants of GANs quantitatively and qualitatively while detecting overfitting and mode collapse.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"CNN-Based Metrics for Performance Evaluation of Generative Adversarial Networks\",\"authors\":\"Adarsh Prasad Behera;Satya Prakash;Siddhant Khanna;Shivangi Nigam;Shekhar Verma\",\"doi\":\"10.1109/TAI.2024.3401650\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this work, we propose two convolutional neural network (CNN) based metrics, classification score (CS) and distribution score (DS), for performance evaluation of generative adversarial networks (GANs). Though GAN-generated images can be evaluated through manual assessment of visual fidelity, it is prolonged, subjective, challenging, tiresome, and can be misleading. Existing quantitative methods are biased toward memory GAN and fail to detect overfitting. CS and DS allow us to experimentally prove that training of GANs is actually guided by the dataset that it improves with every epoch and gets closer to following the distribution of the dataset. Both methods are based on GAN-generated image classification by CNN. CS is the root mean square (rms) value of three different classification techniques, direct classification (DC), indirect classification (IC), and blind classification (BC). It exhibits the degree to which GAN can learn the features and generate fake images similar to real datasets. DS shows the contrast between the mean distribution of GAN-generated data and the real data. It indicates the extent to which GANs can create synthetic images with similar distribution to real datasets. We evaluated CS and DS metrics for different variants of GANs and compared their performances with existing metrics. Results show that CS and DS can evaluate the different variants of GANs quantitatively and qualitatively while detecting overfitting and mode collapse.\",\"PeriodicalId\":73305,\"journal\":{\"name\":\"IEEE transactions on artificial intelligence\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-03-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE transactions on artificial intelligence\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10531154/\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE transactions on artificial intelligence","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/10531154/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
在这项工作中,我们提出了两个基于卷积神经网络(CNN)的指标,即分类得分(CS)和分布得分(DS),用于生成式对抗网络(GAN)的性能评估。虽然可以通过人工评估视觉保真度来评价 GAN 生成的图像,但这种方法耗时长、主观性强、具有挑战性、令人厌烦,而且可能会产生误导。现有的量化方法偏重于记忆 GAN,无法检测到过拟合。CS 和 DS 可以让我们通过实验证明,GAN 的训练实际上是在数据集的指导下进行的,它在每个历时中都会有所改进,并更接近于遵循数据集的分布。这两种方法都是基于由 CNN 生成的 GAN 图像分类。CS 是三种不同分类技术(直接分类 (DC)、间接分类 (IC) 和盲分类 (BC))的均方根值。它显示了 GAN 学习特征并生成与真实数据集相似的假图像的程度。DS 显示了 GAN 生成数据的平均分布与真实数据之间的对比。它表明 GAN 能够生成与真实数据集分布相似的合成图像的程度。我们针对 GAN 的不同变体评估了 CS 和 DS 指标,并将其性能与现有指标进行了比较。结果表明,CS 和 DS 可以定量和定性地评估 GAN 的不同变体,同时检测过度拟合和模式崩溃。
CNN-Based Metrics for Performance Evaluation of Generative Adversarial Networks
In this work, we propose two convolutional neural network (CNN) based metrics, classification score (CS) and distribution score (DS), for performance evaluation of generative adversarial networks (GANs). Though GAN-generated images can be evaluated through manual assessment of visual fidelity, it is prolonged, subjective, challenging, tiresome, and can be misleading. Existing quantitative methods are biased toward memory GAN and fail to detect overfitting. CS and DS allow us to experimentally prove that training of GANs is actually guided by the dataset that it improves with every epoch and gets closer to following the distribution of the dataset. Both methods are based on GAN-generated image classification by CNN. CS is the root mean square (rms) value of three different classification techniques, direct classification (DC), indirect classification (IC), and blind classification (BC). It exhibits the degree to which GAN can learn the features and generate fake images similar to real datasets. DS shows the contrast between the mean distribution of GAN-generated data and the real data. It indicates the extent to which GANs can create synthetic images with similar distribution to real datasets. We evaluated CS and DS metrics for different variants of GANs and compared their performances with existing metrics. Results show that CS and DS can evaluate the different variants of GANs quantitatively and qualitatively while detecting overfitting and mode collapse.