NEWS

2025.8.20

AWL’s Paper Accepted at IEEE ICIP 2025—One of the World’s Largest Conferences in Image Processing and Computer Vision

• NEWS
• プレスリリース

AWL, Inc. (Headquarters: Shinagawa-ku, Tokyo; President & CEO: Muneharu Kitade), a certified Hokkaido-University startup and the winner of the NEDO Chairman’s Award at the “Award for Academic Startups 2024,” is pleased to announce that a research paper submitted by its R&D division has been accepted for presentation at the 2025 IEEE International Conference on Image Processing (IEEE ICIP), to be held in Anchorage, Alaska, USA from September 14 to 18, 2025 (local time).

IEEE ICIP is one of the most prestigious and largest international conferences in the field of image processing and computer vision. It is known for its rigorous selection process, with over a thousand submissions annually from around the world. For IEEE ICIP 2025, approximately 40% of submitted papers were accepted, making AWL’s selection a testament to the global recognition of its technological capabilities and research achievements.

AWL was selected for the FY2023 NEDO “Deep Tech Startup Support Fund / Deep Tech Startup Support Program (DTSU)”. The accepted paper is a result of this initiative and represents a critical technical and strategic milestone for future practical R&D.

■Paper Overview

Title:

GRID-LOGAT: Grid Based Local and Global Area Transcription for Video Question Answering

Authors:

Md Intisar Chowdhury*, Kittinun Aukkapinyo*, Hiroshi Fujimura*, Joo Ann Woo, Wasu Wasusatein, Fadoua Ghourabi (AWL, Inc.)*Co-first authors (equal contribution)

Overview:

AWL specializes in the social implementation of edge AI. This research focuses on video understanding using Visual Language Models (VLMs) that operate on local devices. The paper proposes a novel method called “GRID-LOGAT,” which divides images using grid lines and integrates local and global area information for enhanced processing. This contributes to the advancement of image-language fusion in the era of generative AI.
GRID-LOGAT is designed for local device processing and achieves both privacy protection and high-precision image recognition through detailed analysis based on overlaid grid lines. By minimizing cloud dependency, it also contributes to reduced power consumption.
This technology is aimed at practical applications in security and operational efficiency in stores and facilities. It represents a breakthrough in AWL’s mission to solve social challenges through edge AI.

Related Links:

IEEE ICIP 2025:
https://2025.ieeeicip.org/

Session Details:
https://cmsworkshops.com/ICIP2025/view_session.php?SessionID=1057&bare=1

■About AWL, Inc

AWL, Inc. is a pioneer in edge AI video analytics, with a strong presence in the retail IoT space. Originating from Hokkaido University, the company addresses pressing societal challenges such as aging populations and workforce shortages by delivering scalable, cost-effective AI solutions. It has established a strong partnership with SATUDORA, a leading Hokkaido-based retail chain, to implement these technologies on the ground. AWL has embraced global hiring since its inception, successfully building a diverse engineering team with members from over 20 countries and fostering a culture of inclusivity and innovation akin to that of Silicon Valley.
Established on June 1, 2016. From May 2025, AWL’s Tokyo headquarters will be located in THE CROSS GOTANDA 8F, 2-24-4 Nishi-Gotanda, Shinagawa-ku, while its Sapporo headquarters operates from 1-20 Kita 8-jo Higashi 4-chome, Higashi-ku. The company continues to lead the development and delivery of advanced edge AI video analysis solutions.