Boosting Robustness of Image Matting with Context Assembling and Strong Data Augmentation

Dai, Yutong; Price, Brian; Zhang, He; Shen, Chunhua

doi:10.48550/arxiv.2201.06889

Cited by 1 publication

(1 citation statement)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous methods include affinity-based methods (Aksoy et al, 2018;Levin et al, 2007), sampling-based methods (He et al, 2011;Shahrian et al, 2013), and deep learning based methods (Hou & Liu, 2019;Liu et al, 2021a;Lu et al, 2019;Sun et al, 2021). Besides, there are other methods using different auxiliary inputs, e.g., a background image (Lin et al, 2020;Sengupta et al, 2020), a coarse map (Dai et al, 2022;Yu et al, 2021) or even language descriptions (Li et al, 2023).…”

Section: Image Mattingmentioning

confidence: 99%

Rethinking Portrait Matting with Privacy Preserving

Zhang

et al. 2023

Int J Comput Vis

View full text Add to dashboard Cite

Recently, there has been an increasing concern about the privacy issue raised by identifiable information in machine learning. However, previous portrait matting methods were all based on identifiable images. To fill the gap, we present P3M-10k, which is the first large-scale anonymized benchmark for Privacy-Preserving Portrait Matting (P3M). P3M-10k consists of 10,421 high resolution face-blurred portrait images along with high-quality alpha mattes, which enables us to systematically evaluate both trimap-free and trimap-based matting methods and obtain some useful findings about model generalization ability under the privacy preserving training (PPT) setting. We also present a unified matting model dubbed P3M-Net that is compatible with both CNN and transformer backbones. To further mitigate the cross-domain performance gap issue under the PPT setting, we devise a simple yet effective Copy and Paste strategy (P3M-CP), which borrows facial information from public celebrity images and directs the network to reacquire the face context at both data and feature level. Extensive experiments on P3M-10k and public benchmarks demonstrate the superiority of P3M-Net over state-of-the-art methods and the effectiveness of P3M-CP in improving the cross-domain generalization ability, implying a great significance of P3M for future research and real-world applications. The dataset, code and models are available here (https://github.com/ViTAE-Transformer/P3M-Net).

show abstract

Section: Image Mattingmentioning

confidence: 99%

Rethinking Portrait Matting with Privacy Preserving

Zhang

et al. 2023

Int J Comput Vis

View full text Add to dashboard Cite

show abstract

Boosting Robustness of Image Matting with Context Assembling and Strong Data Augmentation

Cited by 1 publication

References 30 publications

Rethinking Portrait Matting with Privacy Preserving

Rethinking Portrait Matting with Privacy Preserving

Contact Info

Product

Resources

About