Computer Vision news and analysis on Web Pulse: 3339 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
PageWatch.tech has developed a computer vision diff pipeline for web change monitoring that combines SSIM, ORB feature matching, and a Siamese ResNet-18 neural network to handle layout shifts and dynamic content. The app…
Clavicular, the American influencer whose real name is Braden Eric Peters, launched Face Harmony, an AI facial analysis app that scans users' faces to rate symmetry, jawline, and grooming potential, after a highly public…
Geospatial code can execute without errors and produce clean-looking results while being methodologically wrong, a problem amplified by AI coding agents generating GIS workflows from natural-language instructions. Develo…
Nvidia released Medical Physics Simulation, an open-source, GPU-accelerated layer within Isaac for Healthcare that lets developers train and test medical robots using classical physics solvers and the Cosmos-H-Dreams gen…
A developer proposes a model-agnostic prompt architecture for AI video workflows, using a structured intermediate representation with invariants and preferences to ensure consistent results across different generators. T…
BinSentry, a Kitchener, Ont.-based AgTech company, now monitors more than 65,000 animal feed bins in real time across North America and Brazil, up from 4,500 bins four years ago, after recording 100 percent year-over-yea…
NASA's Jet Propulsion Laboratory has successfully demonstrated the first in-orbit use of a vision-language model, Google's Gemma 3, aboard Loft Orbital's YAM-9 satellite. The NAVI-Orbital system achieved 88 percent accur…
Screenpipe offers a new AI workflow for 'perfect memory' by recording screen and audio locally, tracking OS events like app switches and clicks, and combining screenshots with the accessibility tree to understand UI stru…
Sony has unveiled the FX5 compact cinema camera with a 16.6MP full-frame stacked sensor, open-gate recording, and a new BIONZ XR2 processor featuring dedicated AI processing for autofocus and auto white balance. The body…
Cubby Clipboard, a Windows clipboard manager, now indexes image content with OCR-powered search, allowing users to find text inside screenshots from their clipboard history. The tool replaces or enhances the native Win+V…
Software engineer Eric Lu launched Ghost Font, a dynamic typographic tool that uses motion to hide messages from AI models while remaining readable to humans, as part of his ongoing Mixfont project exploring AI and typog…
Google announced a new selfie video feature that lets users sign in to their accounts by capturing a short video of their face, which is compared to a securely stored reference video. The feature, rolling out globally in…
Microsoft released Mage-Flow, a 4-billion-parameter image generation model that aims to match the quality of larger models while being smaller. The model uses a co-design approach with Mage-VAE, a latent tokenizer that r…
Lingverse (formerly Ling AI) raised $29 million to build iKairos, a wearable AI device that can be worn as a pendant or mounted on a table, claiming to turn real-life moments into AI-generated images. CEO Jiawei Gu says …
Packt Publishing has released the third edition of 'Modern Computer Architecture and Organization' by Jim Ledin, adding new chapters on Graphics Processing Units (GPUs) and Large Language Models (LLMs) to address the exp…
Gemini Omni Flash beats Heygen 34.5 to 18.8 in a head-to-head video generation test, winning 3 of 4 tasks with 1 tie and 94% confidence, according to a comparison run by an unnamed tester using GPT-5.4. The test found Ge…
Researchers present SafeGen, a goal-conditioned diffusion framework for generating safety-critical driving scenarios to evaluate vision-language models (VLMs) in autonomous driving systems. SafeGen increases the Judge Ov…
Researchers have developed EGRNet, a lightweight semantic segmentation network with only 0.46 million parameters that achieves 65.28% mean Intersection over Union on the Cityscapes dataset, designed for real-time urban s…
ChronoStitch, a training-free method for composing independently stored visual KV memories, improves long-video question answering by re-basing post-rotary keys onto a global three-axis multimodal RoPE coordinate system …
A new method for detecting AI-generated video operates directly on the compressed bitstream, reading the motion field already written by the codec instead of decoding pixels. The approach achieves a full-length AUC of 0.…