Bridging Photos and Floor Plans with Computer Vision

Cornell University researchers have developed a new computer-vision method, that enables machines to match real-world images with simplified building layouts like floor plans with much greater accuracy. To train and evaluate their approach, the team compiled a large dataset called C3, containing about 90,000 paired photos and floor plans across nearly 600 scenes, with detailed annotations of pixel matches and camera poses. 

By reconstructing scenes in 3D from large internet photo collections and aligning them to publicly available architectural drawings, the dataset teaches models how real images relate to abstract representations. In tests, C3Po reduced matching errors by about 34% compared with earlier methods, suggesting that this multi-modal training could help future vision systems generalize across varied inputs and advance 3D computer vision research.

More information:

https://news.cornell.edu/stories/2025/12/computer-vision-connects-real-world-images-building-layouts

Portable Observatory Monitors Eruptions

Researchers with Istituto Nazionale di Geofisica e Vulcanologia (INGV) deployed a suitcase-sized portable observatory named Setup for the Kinematic Acquisition of Explosive Eruptions (SKATE) on the volcano Stromboli in Italy. The device is equipped with high‐speed cameras, thermal sensors, acoustic sensors and data-acquisition hardware, designed to autonomously monitor explosive eruptions.

SKATE records synchronized thermal, visual and acoustic data, greatly reducing the time scientists must spend in hazardous zones and enabling analysis of more than a thousand explosion events between 2019 and 2024. The detailed high-frame-rate and multiparametric data help volcanologists better understand eruption dynamics and may feed into training libraries for automated warning systems.

More information:

https://spectrum.ieee.org/volcano-monitoring-stromboli-skate

AI Creates Fast Detailed 3D Maps

MIT researchers have built a new AI system that allows robots to create detailed 3D maps of complex environments within seconds. The technology could transform how search-and-rescue robots navigate collapsed mines or disaster sites.

The system combines recent advances in machine learning with classical computer vision principles. It can process an unlimited number of images from a robot’s onboard cameras, generating accurate 3D reconstructions while estimating the robot’s position in real time.

More information:

https://interestingengineering.com/innovation/ai-mapping-system-for-rescue-robots-mit

Interactive Media for Cultural Heritage

Recently, the latest edited book I co-authored with colleagues from CYENS – Centre of Excellence and the University of Cyprus was published by Springer Series on Cultural Computing. The book is entitled ‘Interactive Media for Cultural Heritage’ and presents the full range of interactive media technologies and their applications in Digital Cultural Heritage. It offers a forum for interaction and collaboration among the interactive media and cultural heritage research communities.

The aim of this book is to provide a point of reference for the latest advancements in the different fields of interactive media applied in Digital Cultural Heritage research, ranging from visual data acquisition, classification, analysis and synthesis, 3D modelling and reconstruction, to new forms of interactive media presentation, visualization and immersive experience provision via extended reality, collaborative spaces, serious games and digital storytelling.

More information:

https://link.springer.com/book/10.1007/978-3-031-61018-9

Speed of Light Camera Detection

Scientists have developed a new type of compact camera engineered for computer vision. Developed by scientists from the University of Washington and Princeton University, the prototype uses optics for computing and reduces power consumption.

It also enables the camera to identify objects at the speed of light. Their device also represents a new approach to the field of computer vision, a type of artificial intelligence that allows computers to recognize objects in images and video.

More information:

https://interestingengineering.com/innovation/new-camera-identifies-objects-200x-faster