程序引导的图像操作器
Program-Guided Image Manipulators
摘要：
人类能够在各个层次上构建图像的整体表示，从局部对象到成对关系，再到全局结构。对结构的解释涉及对图像中物体的重复和对称性的推理。在本文中，我们提出了程序引导的图像操纵器（PG-IM），诱导神经符号的程序表示来表示和操纵图像。给定一个图像，PG-IM检测重复的模式，诱导符号程序，并使用由程序引导的神经网络操纵图像。PG-IM从单个图像中学习，利用其内部统计数据。尽管仅接受了图像修补训练，但PG-IM能够在统一的框架中直接进行外推和规则性编辑。大量的实验表明，PG-IM在所有任务上都取得了优异的性能。
英文摘要：
Humans are capable of building holistic representations for images at various levels, from local objects, to pairwise relations, to global structures. The interpretation of structures involves reasoning over repetition and symmetry of the objects in the image. In this paper, we present the Program-Guided Image Manipulator (PG-IM), inducing neuro-symbolic program-like representations to represent and manipulate images. Given an image, PG-IM detects repeated patterns, induces symbolic programs, and manipulates the image using a neural network that is guided by the program. PG-IM learns from a single image, exploiting its internal statistics. Despite trained only on image inpainting, PG-IM is directly capable of extrapolation and regularity editing in a unified framework. Extensive experiments show that PG-IM achieves superior performance on all the tasks.
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。
