The research team of Professor Kang Suk-ju, Department of Electronic Engineering,

Wins the Minister of Trade, Industry and Energy Prize at the '2023 Industry–Academia Project Challenge'

▲ (From left) Professor Kang Suk-ju, Department of Electronic Engineering; Cho Yu-bin, master's student

Cho Yu-bin, a master's student of the Department of Artificial Intelligence (DHE) in the research team of Professor Kang Suk-ju of the Department of Electronic Engineering, won the Minister of Trade, Industry and Energy Prize (Grand Award) at the 2023 Industry–Academia Project Challenge, hosted by the Ministry of Trade, Industry and Energy and the Korea Institute for Advancement of Technology (KIAT). The award ceremony was held on Tuesday 21 November at the AfreecaTV Colosseum in Sangam.

Now in its fourth year, the Industry–Academia Project Challenge is an event in which the various outcomes of the Industrial Innovation Talent Growth Support Program are compared — a competition where master's and doctoral students share and compete on the results of industry–academia projects in which they have researched technical problems arising in industry together with companies.

The team won by proposing a project titled 'Research on Artificial Intelligence Models for Image Segmentation Based on Multimodal In-Vehicle Signals', which was recognized for its excellence. Existing deep learning techniques for autonomous driving image segmentation are limited to predicting only within predefined categories; to address this, the project carried out research on a vision–natural language multimodal segmentation model, improving it so that prediction is possible even for unconstrained language expressions. The team also designed a network for effectively fusing the vision encoder and the language encoder, achieving improved performance.

▲ Photograph from the award of the Minister of Trade, Industry and Energy Prize at the 2023 Industry–Academia Project Challenge

Cho Yu-bin said: "I feel proud that the effort I put into the project seems to have been highly regarded, and I would like to thank my supervisor and everyone who helped with the project. Even after the project, I plan to develop the existing research further in the direction of model lightweighting."

Source: Sogang People https://www.sogang.ac.kr/research/res_01.html