Computer Vision news and analysis on Web Pulse: 3357 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Researchers developed an AI pipeline that digitizes paper ECGs via smartphone photos, achieving 95.51% accuracy in detecting myocardial infarction. The system runs on CPUs in under 30 seconds, enabling remote healthcare …
A developer on Hugging Face highlights the top AI papers of the week, focusing on real-time video generation, interpretable scientific reasoning, long-term memory for robots, and improved benchmarks. Key papers include V…
Researchers have developed MiLSD, a sub-megabyte line segment detector for embedded systems, achieving a leap in performance on the ShanghaiTech Wireframe dataset from 10.6 to 24.1 sAP10 under a one-megabyte memory const…
Robbyant, an embodied AI company within Ant Group, released six models in one week that form a complete full-stack for physical AI, spanning perception, reasoning, action, and world simulation. The stack includes LingBot…
AI-powered surveillance systems will soon track all public and private activities, automatically detecting violations like shoplifting or jaywalking, recording them in government records, and issuing real-time fines and …
A new online game called 'AI or Not' challenges users to distinguish between AI-generated and real images, starting with a street scene. The game presents photorealistic landscapes and asks players to guess whether each …
Researchers introduced CompDiff, a generative model that improves fairness in medical imaging by addressing demographic imbalances through a Hierarchical Conditioner Network. In tests on chest X-rays and fundus images, C…
Meta's AI glasses have sparked widespread controversy over privacy concerns, including unauthorized recording, facial recognition, and human review of footage. Critics highlight issues with consent and tracking, while Me…
AI, drones, and advanced sensors are transforming monitoring in the oil and gas industry, enabling real-time data processing, leak detection, and operational efficiency. This shift reduces costly downtimes and improves s…
Musing Image launched Muse Image, an agentic image model that reasons through prompts, edits photos, composes references, and refines outputs iteratively. The tool is designed for complex tasks like restoration, object r…
VEGAS (Video caption Evaluation via GAze Score) introduces a training-free method to align video captions with viewer gaze data, improving caption-to-video retrieval and personalization. The approach uses real-time gaze …
Researchers introduced WCog-VLA, a dual-level World-Cognitive Vision-Language-Action model that enables autonomous vehicles to transition from reactive to proactive navigation by integrating world cognition with generati…
Researchers have developed VRAIN, a hybrid AI system that combines text-to-image generation with context-aware editing to improve rare-class instance segmentation. The approach achieves up to 4.0 points higher overall Av…
Researchers have developed NamedCurves+, an image editing tool that combines color naming and deep learning to enable intuitive, customizable enhancements. The system uses tone curves and transformer-based spatial depend…
Researchers introduced LEEVLA, a new architecture for vision-language-action models that uses drift-guided dynamic prioritization to focus on task-relevant evidence, outperforming existing methods on VLA benchmarks. The …
Polysense, a Ghent-based startup, raised $10.7 million in seed funding led by Felix Capital to scale its AI quality-control software for food manufacturers. The company's system uses real-time imaging and automated proce…
Estonian retailers are using artificial intelligence in video surveillance systems to analyze shopper behavior and detect shoplifting, with Prisma confirming its use since 2022. The technology raises privacy concerns but…
Researchers propose GRE-Diff, a diffusion-based framework that automates floor plan creation and editing under user constraints, using Gaussian Room Embeddings to model rooms as spatial distributions. The system generate…
Researchers introduced FedTR, a federated learning framework incorporating transfer learning for industrial visual inspection, achieving 95.5% word-level accuracy on homogeneous data and 94.2% on heterogeneous data for l…
Researchers released Infinity-Parser2, a large multimodal model for end-to-end document parsing, along with a 5-million-sample bilingual corpus Infinity-Doc2-5M. The model uses multi-task reinforcement learning across ei…