Computer Vision news and analysis on Web Pulse: 1742 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Vexyn launches a suite of browser-only privacy tools that process data locally using WebGPU AI, with no file uploads, signup, or tracking. The tools include JSON formatting, Base64 encoding, QR generation, PDF merging, i…
Darktable released version 5.6.0 on June 21, 2026, adding an optional AI subsystem with object masking and neural restore modules, disabled by default. The release also includes performance improvements such as OpenCL op…
Researchers at the Technion and the University of Haifa developed an AI tool that retrieves municipal building-permit records and delivers structural information to first responders' mobile devices in under 30 seconds, r…
Code enabling experimental DLSS support for the open-source NVK Vulkan driver was merged into Mesa 26.2-devel on June 19, 2026. The feature, gated behind the NVK_EXPERIMENTAL=dlss environment variable, implements the Vul…
Refloow Photo Studio, a free, open-source desktop photo editor with on-device AI, launched to make professional editing accessible to all skill levels. The app runs locally with no accounts, subscriptions, or data collec…
Kansas City, Missouri, is equipping public buses with facial recognition software to identify banned riders and missing persons, sparking a debate over security versus privacy. The American Civil Liberties Union warns th…
Seeed Studio's reCamera Gimbal has been upgraded with an AI agent that understands natural language, controls hardware, generates code, and plans multi-step tasks, transforming the passive camera into an intelligent assi…
Digital forensics expert Hany Farid warned that ordinary internet users can no longer reliably distinguish AI-generated images, video, or audio from real content, as deepfake content has grown by roughly 900% over the pa…
Digital forensics expert Hany Farid warns that the average person cannot distinguish AI-generated deepfakes from real content, as generative AI has made manipulation effortless and visually indistinguishable. Farid, who …
German startup MicroAGI is offering free apartment cleanings in New York City through its Shift app, with cleaners wearing head-mounted cameras to record first-person footage. The anonymized video is used as training dat…
Zyx Technology is pitching its AI-powered digital twin tools for urban development projects in Uzbekistan, with Chairman Choi Jong-bok presenting the company's solutions at the Korea-Uzbekistan Business Forum. The tools,…
Google's Gemini 1.5 Flash multimodal API can now analyze images and PDFs in Python, returning structured JSON. A new script demonstrates sending a local JPEG and PDF to the model and parsing the response using the google…
A developer launched ImgToolsHub, a curated directory of AI image tools including generators, editors, upscalers, and background removers, to help creative professionals compare and build reliable workflows. The platform…
Google Arts & Culture Lab, in collaboration with artist Stefanie Posavec and the Exploratorium, launched 'See in CMYK', an AI-powered experiment that lets users create custom posters by analyzing photos with Gemini and g…
A developer details their 2026 homelab stack, which runs across four machines including a Mac Mini, Ubuntu mini PC, Raspberry Pi 4, and Raspberry Pi 5 with Hailo 8 AI HAT. The stack uses Traefik for reverse proxying, Tai…
A China-led research team has developed a deep-learning system that automatically detects space hurricanes, a phenomenon in Earth's upper atmosphere that can disrupt satellite signals, radar, and radio communications. Th…
AI agents can automate the entire content creation workflow from research to social posting in a single loop, reducing manual switching between tools. The agentic loop includes research, copywriting, visual generation, a…
Seedance 2 Image to Video defeated Bernini-R Edit Video 17.6 to 13.4 in a head-to-head test of prompt-faithful video generation, with Seedance delivering more precise object reading, action timing, and camera obedience a…
A new scheduling technique called TurboPrefill reduces waiting time for Vision Language Models by nearly half, from 9.0 to 4.6 seconds, without changing model weights or architecture. The optimization, validated on Qwen2…
A self-taught developer with 13 years of experience on the factory floor of a steel-tube manufacturer has built a suite of AI-powered tools to solve real production problems. The tools include PIPEZ, a shift and part-cou…