Tokyo, Japan – June 17, 2026 –　Asia Air Survey Co., Ltd. (Head Office: Shinjuku-ku, Tokyo; President and Representative Director: Megumi Hatakeyama; hereinafter “Asia Air Survey”) is pleased to announce that a paper co-authored by Keisuke Toida of Asia Air Survey and Professor Kazuhiro Hotta of Meijo University has been accepted for presentation at the Image Matching: Local Features & Beyond Workshop, a workshop held in conjunction with CVPR 2026 (Conference on Computer Vision and Pattern Recognition).
CVPR 2026 was held from June 3 to 7, 2026, at the Colorado Convention Center in Denver, Colorado, USA.

The paper, titled “Covisibility-Aware Token Pruning for Scalable Semi-Dense Image Matching,” focuses on image matching, a key technology for identifying corresponding points between images with high accuracy and efficiency.
Image matching supports camera pose estimation and 3D reconstruction from multiple images, and is expected to contribute to a wide range of fields, including SfM, SLAM, visual localization, digital twins, geospatial analysis, infrastructure inspection, construction and civil engineering, robotics, and AR/VR.
Asia Air Survey will continue to combine its expertise in aerial surveying, geospatial information, 3D measurement, infrastructure management, disaster prevention and mitigation, and environmental solutions with AI and computer vision technologies to advance research in image matching, 3D vision, point cloud and image analysis, and digital twin-related technologies.

About CVPR

CVPR (Conference on Computer Vision and Pattern Recognition) is an annual international conference co-sponsored by the IEEE Computer Society and the Computer Vision Foundation (CVF). It is widely regarded as one of the leading conferences in computer vision and pattern recognition, attracting researchers and industry professionals from around the world.
In addition to the main conference, CVPR hosts workshops dedicated to specialized and emerging research areas. Papers accepted to these workshops are selected through each workshop’s review process. Acceptance to the Image Matching: Local Features & Beyond Workshop reflects the relevance and technical contribution of the research to the image matching community.

■ Related Information

• CVPR2026：https://cvpr.thecvf.com/
• Image Matching: Local Features & Beyond WorkShop：https://image-matching-workshop.github.io/
• Asia Air Survey Advanced Technology Research Laboratory：https://www.lab-ajiko.com/
• Hotta Laboratory, Meijo University：https://wwwms.meijo-u.ac.jp/kazuhotta/