Hui-Yi Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Hui-Yi Chen has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Chengju Liu
Tongji University
1 shared publication Horng-Chaung Hsu
Department of Orthopedic Surgery, China Medical University Hospital, No. 2, Yuh-Der Rd, Taichung, Taiwan 404, Republic of China; School of Medicine, China Medical University, No. 91, Hsueh-Shuh Rd, Taichung, Taiwan 40402, Republic of China
1 shared publication- Liuyi Wang
Tongji University
1 shared publication - Qijun Chen
Tongji University
1 shared publication - Ronghao Dang
School of Electronics and Information Engineering, Tongji University, Shanghai, China
1 shared publication - Yen-Jen Chen
School of Medicine, China Medical University, Taichung City 40447, Taiwan
1 shared publication
Latest Publications
- 2 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 33 (7) , 3441-3454
RES-StS: Referring Expression Speaker via Self-Training With Scorer for Goal-Oriented Vision-Language Navigation
It is a rather practical but difficult task to find a specified target object via autonomous exploration based on natural language descriptions in an unstructured environment. Since the human-annotated data is expensive to gather for the goal-oriented vision-language navigation (GVLN) task, the size of the standard dataset is inadequate, which has significantly limited the accuracy of previous techniques. In this work, we aim to improve the robustness and generalization of the navigator by dynamically providing high-quality pseudo-instructions using a proposed RES-StS paradigm. Specifically, we establish a referring expression speaker (RES) to predict descriptive instructions for the given path to the goal object. Based on an environment-and-object fusion (EOF) module, RES derives spatial representations from the input trajectories, which are subsequently encoded by a number of transformer layers. Additionally, given that the quality of the pseudo labels is important for data augmentation while the limited dataset may also hinder RES learning, we propose to equip RES with a more effective generation ability by using the self-training approach. A trajectory-instruction matching scorer (TIMS) network based on contrastive learning is proposed to selectively use rehearsal of prior knowledge. Finally, all network modules in the system are integrated by suggesting a multi-stage training strategy, allowing them to assist one another and thus enhance performance on the GVLN task. Experimental results demonstrate the effectiveness of our approach. Compared with the SOTA methods, our method improves SR, SPL, and RGS by 4.72%, 2.55%, and 3.45% respectively, on the REVERIE dataset, and 4.58%, 3.75% and 3.14% respectively, on the SOON dataset.
- 2 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 33 (7) , 3441-3454
RES-StS: Referring Expression Speaker via Self-Training With Scorer for Goal-Oriented Vision-Language Navigation
It is a rather practical but difficult task to find a specified target object via autonomous exploration based on natural language descriptions in an unstructured environment. Since the human-annotated data is expensive to gather for the goal-oriented vision-language navigation (GVLN) task, the size of the standard dataset is inadequate, which has significantly limited the accuracy of previous techniques. In this work, we aim to improve the robustness and generalization of the navigator by dynamically providing high-quality pseudo-instructions using a proposed RES-StS paradigm. Specifically, we establish a referring expression speaker (RES) to predict descriptive instructions for the given path to the goal object. Based on an environment-and-object fusion (EOF) module, RES derives spatial representations from the input trajectories, which are subsequently encoded by a number of transformer layers. Additionally, given that the quality of the pseudo labels is important for data augmentation while the limited dataset may also hinder RES learning, we propose to equip RES with a more effective generation ability by using the self-training approach. A trajectory-instruction matching scorer (TIMS) network based on contrastive learning is proposed to selectively use rehearsal of prior knowledge. Finally, all network modules in the system are integrated by suggesting a multi-stage training strategy, allowing them to assist one another and thus enhance performance on the GVLN task. Experimental results demonstrate the effectiveness of our approach. Compared with the SOTA methods, our method improves SR, SPL, and RGS by 4.72%, 2.55%, and 3.45% respectively, on the REVERIE dataset, and 4.58%, 3.75% and 3.14% respectively, on the SOON dataset.
- 1 July 2014
- Published by Elsevier in The Spine Journal
- Vol. 14 (7) , 1366
Re: Infection after vertebroplasty or kyphoplasty. A series of nine cases and review of literature. Spine J 13:1809–17
- 1 July 2014
- Published by Elsevier in The Spine Journal
- Vol. 14 (7) , 1366
Re: Infection after vertebroplasty or kyphoplasty. A series of nine cases and review of literature. Spine J 13:1809–17