no code implementations • 31 Dec 2022 • Liguang Zhou, Junjie Hu, Yuhongze Zhou, Tin Lun Lam, Yangsheng Xu
Unbiased scene graph generation (USGG) is a challenging task that requires predicting diverse and heavily imbalanced predicates between objects in an image.
no code implementations • 31 Dec 2022 • Liguang Zhou, Yuhongze Zhou, Xiaonan Qi, Junjie Hu, Tin Lun Lam, Yangsheng Xu
Then, to build multi-scale hierarchical information of input features, we utilize an attention fusion mechanism to aggregate features from multiple layers of the backbone network.
no code implementations • 15 Aug 2022 • Liguang Zhou, Yuhongze Zhou, Tin Lun Lam, Yangsheng Xu
Specifically, we propose to integrate the mixture of experts with a divide and ensemble strategy to remedy the severely long-tailed distribution of predicate classes, which is applicable to the majority of unbiased scene graph generators.
no code implementations • 10 Sep 2021 • Yuhongze Zhou, Liguang Zhou, Tin Lun Lam, Yangsheng Xu
Our MGRConv can be regarded as soft partial convolution and find a trade-off among partial convolution, learnable attention maps, and gated convolution.
no code implementations • 31 Mar 2021 • Yuhongze Zhou, Liguang Zhou, Tin Lun Lam, Yangsheng Xu
This paper presents a semantic-guided automatic natural image matting pipeline with Trimap Generation Network and light-weight non-local attention, which does not need trimap and background as input.
1 code implementation • 4 Jun 2020 • Yuhongze Zhou, Qinjie Xiao
We propose a generative framework based on generative adversarial network (GAN) to enhance facial attractiveness while preserving facial identity and high-fidelity.