Computer Vision news and analysis on Web Pulse: 3351 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
A new arXiv paper (2607.09822v1) introduces a memory-conditioned tool-calling system for camera-first visual agents that uses a three-layer personal visual memory to improve tool-query relevance by 0.47 points (4.21 to 3…
Researchers propose the first diffusion-based unrestricted adversarial attack against 2D LiDAR range-image segmentation, using adversarial guidance from a segmentation loss to generate unrestricted adversarial examples t…
Researchers at arXiv identify compression asymmetry and trajectory binding as key mechanisms in noise-anchored diffusion inversion, introducing Noise-Anchored Reverse Correction (NARC), a training-free inversion primitiv…
Researchers propose a generalized deep non-negative matrix factorization (G-DNMF) method for synthetic aperture radar (SAR) automatic target recognition, which discards the layer-by-layer decomposition strategy to avoid …
Researchers propose a dynamic scene graph attention framework (DSiGAT) that predicts lane-change intention and future trajectory for every vehicle in a local traffic scene, achieving intention prediction accuracies of 90…
Amorepacific's new flagship store Amore Yongsan in Seoul uses AI and robots to create custom foundation, lip products and hair care tailored to each customer, with the Hera Custom Match service offering 335 shades and a …
A developer's guide details how to integrate Claude Code with NanoBanana MCP for image generation and editing directly from the terminal. The integration exposes two tools—nanobanana_generate_image and nanobanana_edit_im…
AI video generation tools can turn static product catalog images into marketing videos, addressing the high cost of traditional video production for e-commerce brands with hundreds or thousands of SKUs. Tools like Runway…
Singapore-based video-generation startup PixVerse has raised $439 million in a Series C extension, pushing its valuation past $2 billion. The company, founded in 2023 by Wang Changhu and Jaden Xie, plans to expand its wo…
The Seventh Circuit Court of Appeals vacated Clearview AI's class-action settlement on July 13, 2026, ruling that the deal giving plaintiffs a 23% equity stake instead of cash suffered from procedural problems. The settl…
Robbyant released six embodied-AI models over four days spanning perception, action, world modeling and video, positioning the LingBot stack as a platform for robotics developers. The public artifacts include LingBot-Vis…
Meta has launched four privacy-eroding features in a month, including using public Instagram photos to train Meta AI, a facial recognition system in Meta Ray-Ban smart glasses, a prototype that continuously records audio…
The Los Angeles Police Department is ending its contract with Flock Safety after an audit found the company's AI-powered license plate cameras generated 161 false stolen-vehicle alerts in two months, a 32.3% error rate t…
A developer released an open-source CLI tool that converts YouTube guitar lesson videos into PDF guitar tabs by downloading the video, using Claude vision to detect tab regions, de-duplicating frames by bar number, and s…
Meta's next smart glasses will reportedly activate the onboard camera for AI features without turning on the camera indicator LED, according to a 9to5Google report. The change, which removes the privacy notification, rai…
Meta's AI-powered Ray-Ban Meta and Oakley Meta smart glasses are transforming sightseeing with hands-free travel features including real-time landmark information, live translation in six languages, and first-person cont…
MIT CSAIL and Toyota Research Institute have developed SceneSmith, a framework using three AI agents powered by GPT-5.2 to build detailed 3D virtual environments for robot training, generating scenes with up to six times…
Path Robotics is deploying AI-driven welding systems that use real-time vision guidance to optimize torch paths and robotic movements, improving precision and efficiency in manufacturing and shipbuilding. The Ohio-based …
MIT engineers have developed an ultrasonic wristband that tracks hand movements in real-time by capturing ultrasound images of wrist muscles and tendons, which an AI algorithm translates into hand and finger positions. T…
MarkTechPost published a tutorial on building a VideoAgent-style multi-agent system for video editing tasks, including intent parsing, graph planning, and tool routing. The system integrates FFmpeg, Whisper transcription…