Articles about "computer vision"

Vivilo, an Italian AI startup, has secured €628,000 in pre-seed funding to expand its proprietary platform for personalised event content generation. The company aims to further develop its recognition technology and strengthen its position in the European market.

Google has advanced its Genie world model to simulate real-world streets by incorporating Street View data. The technology demonstrates how generative AI can recreate complex, realistic environments, potentially impacting industries focused on simulation and autonomous navigation.

This article explores the foundations and contemporary applications of edge detection in computer vision using Python. It provides historical and theoretical context, hands-on code examples, and discusses key algorithms including the Laplacian of Gaussian and Canny edge detector.

Ouster has introduced a new color lidar sensor aimed at replacing traditional camera systems in machine vision applications. The innovation combines lidar's three-dimensional sensing with color data, promising enhanced performance for robotics and autonomous vehicles. The development could influence the design of future AI-powered systems that rely on visual information.

VisioLab, a German AI startup specializing in self-checkout solutions, has raised $11 million in Series A funding to accelerate global expansion. The company’s technology uses computer vision and edge AI to recognize food items, streamlining the checkout process. The new capital will support international growth and product development.

Siemens has significantly advanced the use of artificial intelligence within its manufacturing operations, focusing on predictive maintenance and AI-based visual inspection. These systems reduce equipment downtime and enhance product quality, contributing to measurable improvements in productivity and cost savings. The company's approach reflects the growing maturity of industrial AI deployments in large-scale manufacturing environments.

Apple researchers, in collaboration with Hong Kong University, have presented the LGTM framework to address efficiency challenges in high-resolution 4K 3D scene rendering. The method separates scene geometry and texture, reducing computational demands while improving detail—particularly relevant for Apple's Vision Pro headset. LGTM could set a new benchmark for real-time rendering in consumer devices.

Researchers at MIT have developed an AI-driven method to estimate car emissions in real time at a city-block level using traffic cameras and mobile data. The approach provides high-resolution, privacy-protecting emission maps, offering valuable insights for policymakers on the environmental impact of traffic interventions. The system's effectiveness was demonstrated through studies of Manhattan and Amsterdam.

Researchers at MIT have developed VisiPrint, an AI-powered tool to provide more accurate aesthetic previews for 3D-printed objects. By focusing on color, texture, and material appearance, VisiPrint aims to reduce waste and improve user satisfaction in 3D printing applications.

A collaborative study introduces a computer vision-based system to enhance fish population monitoring, supporting traditional citizen science methods. Researchers built and validated an end-to-end AI pipeline for automated fish counting in Massachusetts rivers, highlighting its scalability and data quality improvements.