Computer Vision news and analysis on Web Pulse: 3337 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Researchers have introduced SAGA (Source Attribution of Generative AI videos), the first comprehensive framework for identifying the specific generative model used to create AI-generated videos, achieving state-of-the-ar…
A developer built hwatu, a browser designed for coding agents to verify UI changes via pixel diffing, achieving a 97.49% match in about 87 ms per verification loop. The tool runs headlessly on Linux using WebKitGTK, supp…
Chrome is integrating AI models that run locally on the user's device, enabling on-device inference for real-time personalization and camera-based interactions without external API calls. This shift allows frontend devel…
Prentis is shifting AI focus from coding to task automation, arguing that the biggest value unlock for LLM agents lies in navigating browsers, interacting with legacy software, and executing multi-step workflows without …
Amazon sellers face listing suppression risk if AI-generated product images cross into deceptive advertising, according to a new compliance guide. The policy requires that the product itself remain an untouched photograp…
A developer built a production-ready OCR service using PaddleOCR-VL-1.6 after finding no trusted provider, achieving a cost of roughly $1 per 1,000 pages. The service undercuts competitors like Mistral OCR 4, which is 4x…
A developer at StoryMine describes the challenge of maintaining character identity and style consistency across 24 pages of an AI-generated picture book. The developer notes that generating each page independently leads …
Creative technologist Tina Tarighian's project '100 Deep Breaths' uses Google's Veo model to generate 100 video studies of AI-generated humans breathing, exploring the unsettling threshold between simulation and reality.…
The Department of Homeland Security's Science & Technology Directorate, in partnership with Cignal through the Silicon Valley Innovation Program, has released a synthetic dataset of 250 X-ray baggage images to improve TS…
A new paper, AREX, introduces recursively self-improving agents that optimize their own research strategies, moving beyond static pipelines. Another study on Diffusion Transformers reveals that text template tokens act a…
Baidu released Unlimited-OCR, an MIT-licensed model that can parse dozens of pages of a PDF in a single forward pass without the GPU memory growth typical of transformer-based OCR. The model uses a novel Reference Slidin…
A new benchmark called ActiveVision reveals that today's multimodal large language models (MLLMs) lack robust active visual observation, with the highest-scoring model, GPT-5.5 at its highest reasoning-effort tier, solvi…
Black Forest Labs and mimic robotics have deployed FLUX-mimic, a video-action model that controls factory robots at Audi with a 101ms full-system reaction time, using a single NVIDIA RTX 5090 GPU. The model, built on FLU…
The Enabot Ebo Max AI robot, a spherical companion droid about the size of a large grapefruit, can patrol homes, answer questions, monitor for falls, and distinguish up to 10 faces using its 4K camera and AI Vision, acco…
Microsoft said on July 23 that Bing Image Creator is now fully powered by its MAI-Image-2.5 model, while PowerPoint and OneDrive also use the model for image editing, reporting up to 84% lower GPU costs than GPT-Image-2 …
A developer reports that lightweight CNNs outperform transformer-based models for client-side background removal in the browser, citing memory constraints that cause transformer attention to exceed WASM limits on mid-ran…
Google and Meta both launched facial recognition features this week, with Google introducing selfie video sign-in for account recovery and Meta launching Facebook Verified to confirm users are human, not AI-generated sca…
Meta announced Facebook Verified, a free optional badge that users can obtain by submitting a facial verification to prove they are human. The process requires users 18 or older to record a short video selfie, which Meta…
YMX Logistics reduced a grocery distributor's yard fleet by 36% using its Autonomous Yard Operating System (YOS), which applies AI to optimize yard operations. The system was honored in FreightWaves' AI Excellence in Sup…
China launched three experimental satellites on July 24 to test artificial intelligence computing technologies in space, using a Lijian-1 Y15 rocket from the Jiuquan Satellite Launch Center. The satellites — Jitian Star …