Computer Vision news and analysis on Web Pulse: 3381 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
In a head-to-head comparison, Imagineart 2.0 Preview decisively defeated Bagel with a score of 26.5 to 18.2, demonstrating superior adherence to complex prompts across three image generation tasks. Bagel produced visuall…
Crit, a new AI-powered design review tool, launches in the terminal for developers to analyze URLs, screenshots, or components with ranked findings and auto-applied fixes. The tool enforces design standards like WCAG con…
Refik Anadol's Dataland, the world's first AI art museum, opened in Los Angeles, using over half a billion pixels and visitor biodata to create immersive experiences that blend nature and technology. The for-profit museu…
A startup called 'Now & Here' launched an Android app three months ago that lets users request near-real-time video of any location and pays nearby people to record it, aiming to solve the trust problem with outdated or …
A large cohort study published in PLOS One found that higher plasma vitamin C levels are independently associated with greater brain volume and stronger default mode network connectivity in older adults, suggesting a pro…
Google Research developed MoGen, an AI model that generates synthetic neuron geometries to improve brain mapping. The model reduced reconstruction errors by 4.4%, potentially saving 157 person-years of manual proofreadin…
Travancore Devaswom Board President K Jayakumar announced an AI-based crowd management system and full cashless digitization for Sabarimala's next pilgrimage season, following Kerala High Court directives. A June 20 work…
A solo founder launched LaunchReel, a tool that converts landing page URLs into platform-optimized social media videos using vision APIs and LLMs to extract content, write captions, and assemble videos with music and voi…
A developer built SnapRegisters, an iOS app that organizes receipts, serial numbers, and warranty terms by product to solve the post-purchase record-keeping problem. The app uses AI to extract model, serial number, and p…
BoardGameBot is an AI-powered Progressive Web App that helps board game players by using OCR to extract text from game components, translating it automatically, and reading it aloud with natural text-to-speech. The app a…
A developer detailed a complete guide to mobile astrophotography stacking using a Motorola Edge 60 Fusion, Fedora Linux, and Siril software. The process involves capturing light, dark, and bias frames, then stacking them…
GeoAxis AI has launched a geolocation tool that can identify the location of any image using AI-powered visual analysis, without relying on GPS or metadata. The tool provides meter-level accuracy in seconds and is design…
Researchers at Rice University's RobotPI Lab released EgoInfinity, a web-scale data engine that generates synthetic video data for training robots to learn actions from video. The tool aims to address data scarcity in ro…
MicroAGI, a German startup, is offering free home cleanings in New York City through its Shift app in exchange for first-person video footage of cleaners at work, which it uses to train and license AI and robot datasets.…
Developer Adam Brown released C2PA Verify, an open-source Android app that reads Content Credentials (C2PA) metadata in photos to reveal whether an image was captured by a camera or generated by AI, aiming to bridge the …
Researchers propose Moebius, a lightweight image inpainting framework with only 0.22B parameters that rivals the performance of 10B-level models like FLUX.1-Fill-Dev. Moebius achieves over 15x faster inference by using a…
Mario Bergeron published Part 4 of his Edge AI Power Benchmarking series on Hackster.io, applying an independent power-measurement methodology to the Axelera Metis accelerator. The post cites vendor-published ResNet-50 v…
Researchers published Sashimi-Bot in npj Robotics, a three-armed robotic system that uses tactile sensing to cut salmon for sashimi. The system achieved 95% accuracy in detecting blade-to-board contact and successfully c…
A developer released PaneTrans, a browser extension for drag-select region translation and OCR on video/canvas, built on Transformers.js with local processing by default. The tool uses an offscreen document for inference…
NVIDIA Research released SpatialClaw, a training-free framework that uses code as the action interface for spatial reasoning in vision-language models. The framework achieves 59.9% average accuracy across 20 benchmarks, …