1 code implementation • 20 Oct 2022 • Yijin Huang, Junyan Lyu, Pujin Cheng, Roger Tam, Xiaoying Tang
Specifically, two saliency-guided learning tasks are employed in SSiT: (1) We conduct saliency-guided contrastive learning based on the momentum contrast, wherein we utilize fundus images' saliency maps to remove trivial patches from the input sequences of the momentum-updated key encoder.
1 code implementation • 27 Jul 2022 • Junyan Lyu, Yiqi Zhang, Yijin Huang, Li Lin, Pujin Cheng, Xiaoying Tang
To address this issue, we propose a data manipulation based domain generalization method, called Automated Augmentation for Domain Generalization (AADG).
2 code implementations • 27 Oct 2021 • Yijin Huang, Li Lin, Pujin Cheng, Junyan Lyu, Roger Tam, Xiaoying Tang
To identify the key components in a standard deep learning framework (ResNet-50) for DR grading, we systematically analyze the impact of several major components.
1 code implementation • 29 Sep 2021 • Huilin Yang, Junyan Lyu, Pujin Cheng, Roger Tam, Xiaoying Tang
We innovatively propose a flexible and consistent cross-annotation face alignment framework, LDDMM-Face, the key contribution of which is a deformation layer that naturally embeds facial geometry in a diffeomorphic way.
no code implementations • 2 Aug 2021 • Huilin Yang, Junyan Lyu, Pujin Cheng, Xiaoying Tang
Instead of predicting facial landmarks via heatmap or coordinate regression, we formulate this task in a diffeomorphic registration manner and predict momenta that uniquely parameterize the deformation between initial boundary and true boundary, and then perform large deformation diffeomorphic metric mapping (LDDMM) simultaneously for curve and landmark to localize the facial landmarks.
2 code implementations • 17 Jul 2021 • Yijin Huang, Li Lin, Pujin Cheng, Junyan Lyu, Xiaoying Tang
Instead of taking entire images as the input in the common contrastive learning scheme, lesion patches are employed to encourage the feature extractor to learn representations that are highly discriminative for DR grading.
1 code implementation • 10 Jul 2021 • Li Lin, Zhonghua Wang, Jiewei Wu, Yijin Huang, Junyan Lyu, Pujin Cheng, Jiong Wu, Xiaoying Tang
Moreover, both low-level and high-level features from the aforementioned three branches, including shape, size, boundary, and signed directional distance map of FAZ, are fused hierarchically with features from the diagnostic classifier.