South Korea’s Naver (KRX: 035420) has secured acceptance of 23 research papers at ECCV (European Conference on Computer Vision) 2026, widely regarded as one of the most authoritative academic gatherings in the field of computer vision. The accepted work spans the full spectrum of spatial intelligence\u2014from technology that reconstructs indoor spaces in 3D from a single photograph to models capable of learning entire cities to enable virtual exploration.<\/p>\n
Naver announced on August 28 that 23 papers submitted by research organizations under Team Naver had been accepted to ECCV 2026. ECCV is an international AI conference covering computer vision topics such as video and image processing, held biennially and now in its 19th edition. This year’s conference will take place in Sweden from September 8 to 12.<\/p>\n
Technology organizations across Team Naver\u2014including Naver Labs, Naver Labs Europe, Naver AI Lab, and Naver Cloud\u2014will present their accepted papers on-site. The research topics encompass the breadth of computer vision, including 3D spatial recognition and reconstruction, image and video learning models, and robot navigation simulation. Four of the papers were selected for the “ECCV Spotlight,” a distinction granted to only approximately 1\u20132% of all submissions.<\/p>\n
Naver Labs Europe will unveil “BLASt3R,” a follow-up to the 3D reconstruction technology “DUSt3R” first released in 2023. While the original DUSt3R reconstructs entire spaces in three dimensions from just one or two photographs, BLASt3R incorporates SLAM (Simultaneous Localization and Mapping) techniques that allow robots to build surrounding maps in real time as they move. This integration dramatically improves speed and accuracy, enabling 3D reconstruction even in on-device environments without server connectivity.<\/p>\n
Naver Labs will present “InSpace,” an indoor 3D spatial generation study that reconstructs room structures and objects from a single 360-degree panoramic image. The team will also unveil technology that leverages vision-language models (VLMs) during autonomous navigation to recognize objects not encountered during training and reflect them in real time on bird’s-eye-view (BEV) maps.<\/p>\n
The “Seoul World Model,” a joint industry-academic research project between Naver AI Lab and KAIST, was also accepted as a regular paper. The model leverages Naver’s accumulated spatial data\u2014including panoramic imagery of the entire city of Seoul\u2014to generate long-range video footage set in the real Seoul. It has potential applications in map-based virtual exploration, digital twins, and autonomous driving and robotics simulation.<\/p>\n
Naver Cloud will present research on accurately recognizing object motion in video, as well as “SpatialBoost,” a study that enhances depth perception performance by adding text-based reasoning to spatial imagery.<\/p>\n
The four papers selected for the Spotlight at this year’s conference include InSpace and the Seoul World Model. Team Naver has maintained a track record of double-digit paper acceptances at major international computer vision conferences each year, including ECCV, CVPR, and ICCV.<\/p>\n
A Naver representative said, “Team Naver will continue to pursue active research and development to secure next-generation AI and spatial intelligence technologies, and will integrate them into a variety of services to deliver technology experiences that users can readily feel in their daily lives.”<\/p>\n","protected":false},"excerpt":{"rendered":"South Korea’s Naver (KRX: 035420) has secured acceptance of 23 research papers at ECCV (European Conference on Computer…\n","protected":false},"author":2,"featured_media":135394,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[623],"tags":[68551,2758,57770,68734,6944,657,68733,944,2757,34601,44849],"class_list":["post-135393","post","type-post","status-publish","format-standard","has-post-thumbnail","category-naver","tag-blast3r","tag-dust3r","tag-eccv-2026","tag-inspace","tag-kaist","tag-naver","tag-naver-ai-lab","tag-naver-cloud","tag-naver-labs","tag-naver-labs-europe","tag-seoul-world-model"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/posts\/135393","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/comments?post=135393"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/posts\/135393\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/media\/135394"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/media?parent=135393"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/categories?post=135393"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/tags?post=135393"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}