arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2012.10043cs.CVcs.AIcs.LG
收藏

内容掩蔽损失:强化学习绘画代理中的类人画笔笔划规划

Content Masked Loss: Human-Like Brush Stroke Planning in a Reinforcement Learning Painting Agent

Carnegie Mellon University(卡内基梅隆大学)

Peter Schaldenbrand, Jean Oh

摘要:

大多数强化学习绘画代理的目标是最小化目标图像和画布之间的损失。人类画家的艺术性强调目标图像的重要特征,而不是简单地复制它(DiPaola 2007)。在RL绘画模型中使用对抗性或L2损失,尽管其最终输出通常是精细的工作,但产生的笔划序列与人类产生的笔划序列大不相同,因为模型不知道目标图像中的抽象特征。为了在不使用昂贵的人类数据的情况下增加模型的人性化规划,我们引入了一个新的损失函数与模型的奖励函数一起使用:内容掩蔽损失。在机器人绘画的上下文中,内容掩蔽丢失采用目标检测模型来提取特征,这些特征用于将较高权重分配给人类将发现对于识别内容重要的画布区域。基于332名人类评估者的结果表明,由我们的内容掩蔽模型产生的数字绘画在笔划序列中比现有方法更早地显示可检测的主题,而不会影响最终绘画的质量。我们的代码可在https://github.com/pschaldenbrand/ContentMaskedLoss上获得。

英文摘要:

The objective of most Reinforcement Learning painting agents is to minimize the loss between a target image and the paint canvas. Human painter artistry emphasizes important features of the target image rather than simply reproducing it (DiPaola 2007). Using adversarial or L2 losses in the RL painting models, although its final output is generally a work of finesse, produces a stroke sequence that is vastly different from that which a human would produce since the model does not have knowledge about the abstract features in the target image. In order to increase the human-like planning of the model without the use of expensive human data, we introduce a new loss function for use with the model's reward function: Content Masked Loss. In the context of robot painting, Content Masked Loss employs an object detection model to extract features which are used to assign higher weight to regions of the canvas that a human would find important for recognizing content. The results, based on 332 human evaluators, show that the digital paintings produced by our Content Masked model show detectable subject matter earlier in the stroke sequence than existing methods without compromising on the quality of the final painting. Our code is available at https://github.com/pschaldenbrand/ContentMaskedLoss.

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。

↑