arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.00818cs.CVcs.AI
收藏

改进模型的焦点可提高基于深度学习的合成人脸检测器的性能

Improving Model's Focus Improves Performance of Deep Learning-Based Synthetic Face Detectors

University of Notre Dame(圣母大学)

Jacob Piland, Adam Czajka, Christopher Sweet

摘要:

基于深度学习的模型在通过将人类感知纳入训练策略来指导“在哪里看”之后,可以更好地推广到未知的数据样本。我们观察到,与为没有人类感知智能的模型训练计算的显着性熵相比,以这种方式训练的模型显着性熵较低。因此,问题是:通过降低模型的类激活图的熵,进一步增加模型的焦点,有助于进一步提高性能吗?在本文中,我们提出并评估了几个基于熵的新的损失函数组件控制模型的重点,涵盖了整个范围的水平,这样的控制,从没有到其“积极的”最小化。我们使用合成人脸检测问题表明,通过降低熵来提高模型的焦点,可以使模型在开集场景中表现得更好,其中测试样本由未知的生成模型合成。我们还表明,当模型的损失函数混合了三个方面时,可以获得最佳性能:常规分类,模型焦点的低熵和人类指导的显着性。

英文摘要:

Deep learning-based models generalize better to unknown data samples after being guided "where to look" by incorporating human perception into training strategies. We made an observation that the entropy of the model's salience trained in that way is lower when compared to salience entropy computed for models training without human perceptual intelligence. Thus the question: does further increase of model's focus, by lowering the entropy of model's class activation map, help in further increasing the performance? In this paper we propose and evaluate several entropy-based new loss function components controlling the model's focus, covering the full range of the level of such control, from none to its "aggressive" minimization. We show, using a problem of synthetic face detection, that improving the model's focus, through lowering entropy, leads to models that perform better in an open-set scenario, in which the test samples are synthesized by unknown generative models. We also show that optimal performance is obtained when the model's loss function blends three aspects: regular classification, low-entropy of the model's focus, and human-guided saliency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑