When compared to the image classification models, black-box adversarial attacks against video classification models have been largely understudied. This could be possible because, with video, the temporal dimension poses significant additional challenges in gradient estimation. Query-efficient black-box attacks rely on effectively estimated gradients towards maximizing the probability of misclassifying the target video. In this work, we demonstrate that such effective gradients can be searched for by parameterizing the temporal structure of the search space with geometric transformations. Specifically, we design a novel iterative algorithm Geometric TRAnsformed Perturbations (GEO-TRAP), for attacking video classification models. GEO-TRAP employs standard geometric transformation operations to reduce the search space for effective gradients into searching for a small group of parameters that define these operations. This group of parameters describes the geometric progression of gradients, resulting in a reduced and structured search space. Our algorithm inherently leads to successful perturbations with surprisingly few queries. For example, adversarial examples generated from GEO-TRAP have better attack success rates with ~73.55% fewer queries compared to the state-of-the-art method for video adversarial attacks on the widely used Jester dataset. Overall, our algorithm exposes vulnerabilities of diverse video classification models and achieves new state-of-the-art results under black-box settings on two large datasets. Code is available here: https://github.com/sli057/Geo-TRAP
翻译:与图像分类模型相比,对视频分类模型的黑盒子对抗性攻击基本上未得到充分研究。这可能是可能的,因为通过视频,时间因素给梯度估计带来了更大的额外挑战。快速高效的黑盒子攻击依靠有效估计梯度来最大限度地扩大目标视频分类的概率。在这项工作中,我们证明,通过将搜索空间的时间结构与几何变参数比较,可以搜索这些有效的梯度。具体地说,我们设计了一个新的迭代代代算法对视频分类模型进行几何马路成形的地形变形(GEO-TRAP),以攻击视频分类模型。GEO-TRAP使用标准的几何几何几度变形转换操作来减少有效梯度的搜索空间,以寻找界定这些操作的一小组参数。这组参数描述了梯度的几何进展,导致搜索空间缩小和结构化。我们的算法必然导致扰动成功。例如,GEOEO-TRAP生成的对抗性梯度变数分析模型(GO-Ar-Arthal-logislational angal angal laction laction laftal laft laction laction laft laction laction laft laction laft labs laction s brogs laft laft lactions labs labs laft labs labs labs lactions labs labs labs laction labs labal labal labs labs labs labal labal labal labal labs labs labalds labs labal labal labalds labal lads labal labal labal labal labaldaldaldaldal lads ladalds labal labal ladal ladal lads lads labal labal ladal ladal ladal labal ladal