VALSE 论文速览 第222期：In-Context Matting

论文题目：

In-Context Matting

作者列表：

He Guo (Huazhong University of Science and Technology), Zixuan Ye (Huazhong University of Science and Technology), Zhiguo Cao (Huazhong University of Science and Technology), Hao Lu (Huazhong University of Science and Technology)

B站观看网址：

https://www.bilibili.com/video/BV1zio7YhEt2/

论文摘要：

We introduce in-context matting, a novel task setting of image matting. Given a reference image of a certain foreground and guided priors such as points, scribbles, and masks, in-context matting enables automatic alpha estimation on a batch of target images of the same foreground category, without additional auxiliary input.

This setting marries good performance in auxiliary input-based matting and ease of use in automatic matting, which finds a good trade-off between customization and automation.

To overcome the key challenge of accurate foreground matching, we introduce IconMatting, an in-context matting model built upon a pre-trained text-to-image diffusion model. Conditioned on inter- and intra-similarity matching, IconMatting can make full use of reference context to generate accurate target alpha mattes. To benchmark the task, we also introduce a novel testing dataset ICM-57, covering 57 groups of real-world images.

Quantitative and qualitative results on the ICM-57 testing set show that IconMatting rivals the accuracy of trimap-based matting while retaining the automation level akin to automatic matting.

参考文献：

[1] He Guo, Zixuan Ye, Zhiguo Cao, Hao Lu, “In-Context Matting,” in Proceeding of IEEE Computer Vision and Pattern Recognition (CVPR 2024), Seattle, WA, June 2024.

论文链接：

https://openaccess.thecvf.com/content/CVPR2024/papers/Guo_In-Context_Matting_CVPR_2024_paper.pdf

视频讲者简介：

He Guo is a master student at Huazhong University of Science and Technology. He received his B.S. degree from the School of Artificial Intelligence and Automation at Huazhong University of Science and Technology. His research interests include computer vision. Recently, his research is focused on the application of generative models.

个人主页：

https://github.com/guohe369

特别鸣谢本次论文速览主要组织者：

月度轮值AC：杨旭 (西安电子科技大学)