We present a novel approach to tackle domain adaptation between synthetic and real data. Instead, of employing "blind" domain randomization, i.e., augmenting synthetic renderings with random backgrounds or changing illumination and colorization, we leverage the task network as its own adversarial guide toward useful augmentations that maximize the uncertainty of the output. To this end, we design a min-max optimization scheme where a given task competes against a special deception network to minimize the task error subject to the specific constraints enforced by the deceiver. The deception network samples from a family of differentiable pixel-level perturbations and exploits the task architecture to find the most destructive augmentations. Unlike GAN-based approaches that require unlabeled data from the target domain, our method achieves robust mappings that scale well to multiple target distributions from source data alone. We apply our framework to the tasks of digit recognition on enhanced MNIST variants, classification and object pose estimation on the Cropped LineMOD dataset as well as semantic segmentation on the Cityscapes dataset and compare it to a number of domain adaptation approaches, thereby demonstrating similar results with superior generalization capabilities.
翻译:我们提出了一种新颖的方法来解决合成数据和真实数据之间的领域适应问题。 相反,我们利用“盲”域随机化,即增加具有随机背景的合成图像,或改变照明和色彩化,利用任务网络作为其自身的对抗性指南,实现有用的增强,使输出的不确定性最大化。为此,我们设计了一个微量最大优化方案,在特定任务与特殊欺骗网络竞争的情况下,尽量减少任务错误,但受欺骗者施加的具体限制。来自不同像素级扰动的欺骗网络样本,利用任务结构寻找最具破坏性的增强力。与GAN型方法不同,因为GAN型方法需要目标领域未加标记的数据,我们的方法实现了强有力的绘图,不仅从源数据中得出多个目标分布。我们运用了自己的框架,对强化的MNIST变异、分类和对象进行数字化识别,对裁剪的LineMOD数据集进行了估计,并用语义分解对市立数据集进行了估算,并将其与一些域适应方法进行了比较,从而展示了类似的结果,同时展示了高超能力。