Computer Vision news and analysis on Web Pulse: 1841 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Google released Gemma 4 12B, a multimodal AI model with an encoder-free architecture that processes images and audio directly without separate encoders, enabling on-device agentic workflows on standard laptops. The 12-bi…
The Ray-Ban Meta Wayfarer (Gen 2) smart glasses offer convenient hands-free photo and video capture, quality speakers for music and calls, and useful features for motorcycle riders, but suffer from limited battery life, …
An injured teenage survivor of a January 2025 school shooting in Nashville, Tennessee has sued Omnilert, the manufacturer of an AI gun detection system that failed to detect the handgun used in the attack. The lawsuit, f…
A developer implemented a perception system for game AI agents, adapted from Crytek's Target Tracks system. The system uses sight and awareness thresholds to track targets and adjust agent behavior, including turning tow…
A new benchmark dataset, the Social Media Robustness Benchmark, evaluates how deepfake detectors perform on images re-encoded by social media platforms like Instagram, Facebook, TikTok, and X. The dataset contains 2,400 …
Developer Anil Madhavapeddy released TESSERA v1.1 and GeoTessera 0.9, improving temporal embeddings and S3+Zarr support, while also advancing io-uring libraries for OCaml. The updates address performance issues and expan…
A new tutorial series guides developers through building a 3D scene editor in Bevy 0.19, inspired by Blender. The first episode covers setting up a basic 3D scene with a camera, light, and cube on an infinite grid, using…
Topaz Video released version 1.6.1, a maintenance update for Windows and macOS, on June 6, 2026. The 318 MB update fixes a crash caused by odd-sized video outputs and warns users that outputting above UHD resolution can …
GPT Image 2 API defeated AuraFlow in a direct comparison, winning every task with a final score of 27.5 to 18.6. The API demonstrated superior precision by accurately rendering specific prompt details, such as a goosenec…
Punjab Safe Cities Authority (PSCA) identified and traced more than 170 vehicles with fake, cloned, or tampered number plates over the past month using its AI-powered Safe City surveillance network. The system's automate…
Ideogram 4, a 9.3-billion-parameter text-to-image model from a team of former Google Brain researchers, topped the open-weight leaderboard for design quality, winning nearly half of blind typography tests against models …
Neatleaf's "Spyder" autonomous robotic scanner has been deployed in PURPLEFARM's 86,000-square-foot cannabis cultivation facility in Fredericton, New Brunswick, where it continuously monitors plant health metrics includi…
Researchers introduced CoopSR, the first benchmark for multi-robot cooperative spatial reasoning, along with the EgoTeam dataset containing 114,227 question-answer pairs across 19 question types. The team also proposed S…
FIFA has integrated artificial intelligence into the official match ball for this year's World Cup, enabling real-time tracking and data-driven officiating. The technology, demonstrated at FIFA headquarters in Zurich, al…
StructOCR launched a new API that uses proprietary AI models to parse data from global passports, invoices, receipts, and shipping containers into structured JSON format. The tool aims to automate KYC, accounts payable, …
OpenCV 5.0 was released today as a major update to the open-source computer vision library, featuring a rewritten deep neural network engine and built-in support for large language and vision language models. The new ver…
Chinese labs previously involved in selling fentanyl precursors have shifted to manufacturing and selling peptides, a gray market now valued at over $100 million annually and growing, according to a new Chainalysis analy…
ESA researchers David O'Ryan and Pablo Gómez used an AI tool called AnomalyMatch to scan nearly 100 million image cutouts from the Hubble Legacy Archive, identifying roughly 1,400 anomalous objects in about two and a hal…
Alibaba's Qwen team released Qwen3.7-Plus, a multimodal agent model that integrates visual perception, GUI operation, and coding into a single autonomous loop. In a demonstration, an agent built on the model independentl…
A blind test of 240 images by 10 professional designers found that Ideogram 4.0 won typography in 47.9% of matchups, significantly outperforming Gemini 3.1 at 30% and FLUX.2 Max and Grok Imagine at roughly 15% each. On t…