UIT students had their scientific paper accepted for publication in a Q1-ranked journal.

Fri, 20/10/2023 - 23:14

Article Title: "Improving Human-Object Interaction with Auxiliary Semantic Information and Enhanced Instance Representation"

Link to the Article:

Authors:

Le Viet Thinh - 20520781 - Computer Science - Co-author

Van Nguyen Ngoc Huyen - 20521424 - Computer Science - Co-author

Supervising Teacher: Dr. Nguyễn Tấn Trần Minh Khang

Article Summary:

Human–Object Interaction (HOI) detection has garnered considerable attention among computer vision researchers as it involves identifying and describing actions between humans and objects. Numerous approaches, such as sequential and end-to-end methods, have been proposed to tackle this problem, with a recent focus on exploring end-to-end systems. This study presents an enhanced end-to-end transformer-based human–object detector based on HOTR, which introduces three improvements. The proposed model improves instance representation through a simple yet effective mechanism, utilizes semantic information to provide contextual understanding and additional knowledge, and incorporates a cross-attention mechanism for fusing multi-level high-level feature maps within the Transformer architecture. Experimental results demonstrate significant performance gains over the baseline HOTR model, making it competitive with other state-of-the-art models on two widely-used datasets: V-COCO and HICO-DET.

We extend our gratitude to Dr. Nguyễn Tấn Trần Minh Khang, Vice Rector of the University of Information Technology, for his dedicated guidance, pinpointing our limitations during the research and publication of this international scientific article.

Pattern Recognition Letters is an international journal specialized in the field of Pattern Recognition, ranked Q1 with an Impact Factor of 5.1 (according to ScienceDirect), standing 14th globally in the field of Computer Vision and Pattern Recognition and 17th globally in the field of Signal Processing (according to Scimago Institutions Rankings).