no code implementations • 19 Mar 2022 • Yiting Li, Haiyue Zhu, Xijia Feng, Zilong Cheng, Jun Ma, Cheng Xiang, Prahlad Vadakkepat, Tong Heng Lee
Specifically, in the \textbf{Implanting} step, we propose to mimic the data distribution of novel classes with the assistance of data-abundant base set, so that a model could learn semantically-rich features that are beneficial for discriminating between the base and other unseen classes.
no code implementations • 23 Sep 2021 • Yiting Li, Haiyue Zhu, Jun Ma, Chek Sing Teo, Cheng Xiang, Prahlad Vadakkepat, Tong Heng Lee
We conduct experiments on both Pascal VOC and MS-COCO, which demonstrate that our method can effectively solve the problem of incremental few-shot detection and significantly improve the detection accuracy on both base and novel classes.
no code implementations • CVPR 2021 • Yiting Li, Haiyue Zhu, Yu Cheng, Wenxin Wang, Chek Sing Teo, Cheng Xiang, Prahlad Vadakkepat, Tong Heng Lee
The failure modes of FSOD are investigated that the performance degradation is mainly due to the classification incapability (false positives), which motivates us to address it from a novel aspect of hard example mining.
no code implementations • 20 Aug 2020 • Haiyue Zhu, Yiting Li, Fengjun Bai, Wenjie Chen, Xiaocong Li, Jun Ma, Chek Sing Teo, Pey Yuen Tao, Wei. Lin
The proposed grasping detection network specially provides a prediction uncertainty estimation mechanism by leveraging on Feature Pyramid Network (FPN), and the mean-teacher semi-supervised learning utilizes such uncertainty information to emphasizing the consistency loss only for those unlabelled data with high confidence, which we referred it as the confidence-driven mean teacher.
no code implementations • 6 May 2020 • Yiting Li, Haiyue Zhu, Sichao Tian, Fan Feng, Jun Ma, Chek Sing Teo, Cheng Xiang, Prahlad Vadakkepat, Tong Heng Lee
Incremental few-shot learning is highly expected for practical robotics applications.