AURA：使用随机化输入采样进行物体移除的自动掩码生成器
AURA : Automatic Mask Generator using Randomized Input Sampling for Object Removal
AI总结：
本文提出受XAI启发的AURA自动掩码生成器，通过随机输入采样、重要性图和裁判模块为现成图像修复网络选择更优物体移除掩码，并提出FID*与U-IDS*指标。
摘要：
图像修复任务的目标是以视觉上合理的方式填充图像中的缺失区域。近年来，基于深度学习的图像修复网络已经取得了出色的结果，其中一些方法通过对图像中不需要的物体进行掩码，将其模型用作物体移除器。然而，以往工作在试图利用其网络更好地移除物体时，对输入掩码的重要性关注较少。本文聚焦于生成输入掩码，以便利用现成的图像修复网络更好地移除物体。我们提出了一种受可解释人工智能（XAI）方法启发的自动掩码生成器，其输出相比语义分割掩码能够更好地移除物体。所提方法利用随机采样的输入掩码，以及由这些随机掩码得到的补全图像的定量估计分数，生成重要性图。输出掩码由裁判模块在根据重要性图生成的候选掩码中选出。我们设计该裁判模块，用于定量估计物体移除结果的质量。此外，我们通过实验发现，以往报告物体移除结果的工作所使用的评估方法并不适合估计物体移除器的性能。因此，我们提出了新的评估指标（FID* 和 U-IDS*），以恰当评估物体移除器的质量。实验证实，与由语义分割图生成的掩码相比，我们的方法在移除目标类别物体方面表现出更好的性能，并且所提出的两个指标能够作出与人类一致的判断。
英文摘要：
The objective of the image inpainting task is to fill missing regions of an image in a visually plausible way. Recently, deep-learning-based image inpainting networks have generated outstanding results, and some utilize their models as object removers by masking unwanted objects in an image. However, while trying to better remove objects using their networks, the previous works pay less attention to the importance of the input mask. In this paper, we focus on generating the input mask to better remove objects using the off-the-shelf image inpainting network. We propose an automatic mask generator inspired by the explainable AI (XAI) method, whose output can better remove objects than a semantic segmentation mask. The proposed method generates an importance map using randomly sampled input masks and quantitatively estimated scores of the completed images obtained from the random masks. The output mask is selected by a judge module among the candidate masks which are generated from the importance map. We design the judge module to quantitatively estimate the quality of the object removal results. In addition, we empirically find that the evaluation methods used in the previous works reporting object removal results are not appropriate for estimating the performance of an object remover. Therefore, we propose new evaluation metrics (FID$^*$ and U-IDS$^*$) to properly evaluate the quality of object removers. Experiments confirm that our method shows better performance in removing target class objects than the masks generated from the semantic segmentation maps, and the two proposed metrics make judgments consistent with humans.
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。
