当前位置: X-MOL 学术J. Comput. Sci. Tech. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Natural Image Matting with Attended Global Context
Journal of Computer Science and Technology ( IF 1.9 ) Pub Date : 2023-05-30 , DOI: 10.1007/s11390-022-1690-z
Yi-Yi Zhang , Li Niu , Yasushi Makihara , Jian-Fu Zhang , Wei-Jie Zhao , Yasushi Yagi , Li-Qing Zhang

Image matting is to estimate the opacity of foreground objects from an image. A few deep learning based methods have been proposed for image matting and perform well in capturing spatially close information. However, these methods fail to capture global contextual information, which has been proved essential in improving matting performance. This is because a matting image may be up to several megapixels, which is too big for a learning-based network to capture global contextual information due to the limit size of a receptive field. Although uniformly downsampling the matting image can alleviate this problem, it may result in the degradation of matting performance. To solve this problem, we introduce a natural image matting with the attended global context method to extract global contextual information from the whole image, and to condense them into a suitable size for learning-based network. Specifically, we first leverage a deformable sampling layer to obtain condensed foreground and background attended images respectively. Then, we utilize a contextual attention layer to extract information related to unknown regions from condensed foreground and background images generated by a deformable sampling layer. Besides, our network predicts a background as well as the alpha matte to obtain more purified foreground, which contributes to better qualitative performance in composition. Comprehensive experiments show that our method achieves competitive performance on both Composition-1k and the alphamatting.com benchmark quantitatively and qualitatively.



中文翻译:

具有参与的全局背景的自然图像抠图

图像抠图是估计图像中前景物体的不透明度。人们已经提出了一些基于深度学习的方法用于图像抠图,并且在捕获空间紧密信息方面表现良好。然而,这些方法无法捕获全局上下文信息,而这已被证明对于提高抠图性能至关重要。这是因为抠图图像可能高达几百万像素,由于接受域大小的限制,对于基于学习的网络来说太大了,无法捕获全局上下文信息。虽然对抠图图像进行均匀下采样可以缓解这个问题,但它可能会导致抠图性能下降。为了解决这个问题,我们引入了一种自然图像抠图,使用有人参与的全局上下文方法从整个图像中提取全局上下文信息,并将它们压缩成适合基于学习的网络的大小。具体来说,我们首先利用可变形采样层分别获得压缩的前景和背景图像。然后,我们利用上下文注意层从可变形采样层生成的压缩前景和背景图像中提取与未知区域相关的信息。此外,我们的网络还预测背景和 alpha 遮罩以获得更纯净的前景,这有助于提高构图的定性性能。综合实验表明,我们的方法在 Composition-1k 和 alphamatting.com 基准上定量和定性地实现了具有竞争力的性能。我们首先利用可变形采样层分别获得压缩的前景和背景图像。然后,我们利用上下文注意层从可变形采样层生成的压缩前景和背景图像中提取与未知区域相关的信息。此外,我们的网络还预测背景和 alpha 遮罩以获得更纯净的前景,这有助于提高构图的定性性能。综合实验表明,我们的方法在 Composition-1k 和 alphamatting.com 基准测试上在定量和定性方面都取得了有竞争力的性能。我们首先利用可变形采样层分别获得压缩的前景和背景图像。然后,我们利用上下文注意层从可变形采样层生成的压缩前景和背景图像中提取与未知区域相关的信息。此外,我们的网络还预测背景和 alpha 遮罩以获得更纯净的前景,这有助于提高构图的定性性能。综合实验表明,我们的方法在 Composition-1k 和 alphamatting.com 基准上定量和定性地实现了具有竞争力的性能。我们利用上下文注意层从可变形采样层生成的压缩前景和背景图像中提取与未知区域相关的信息。此外,我们的网络还预测背景和 alpha 遮罩以获得更纯净的前景,这有助于提高构图的定性性能。综合实验表明,我们的方法在 Composition-1k 和 alphamatting.com 基准上定量和定性地实现了具有竞争力的性能。我们利用上下文注意层从可变形采样层生成的压缩前景和背景图像中提取与未知区域相关的信息。此外,我们的网络还预测背景和 alpha 遮罩以获得更纯净的前景,这有助于提高构图的定性性能。综合实验表明,我们的方法在 Composition-1k 和 alphamatting.com 基准测试上在定量和定性方面都取得了有竞争力的性能。

更新日期:2023-05-30
down
wechat
bug