Computer Vision news and analysis on Web Pulse: 3330 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Apple has three new smart home products nearly ready to launch, according to a Bloomberg report. The devices include a new Apple TV set-top box and an updated HomePod mini with faster chips for Siri AI, expected as soon …
Vision-language models (VLMs) fail at price estimation because they over-index on surrounding context rather than analyzing material properties, according to a technical analysis. When a necklace was paired with an expen…
A study by researcher Brianne Lee found that six multimodal large language models priced a $2.43 Temu necklace at $19 to $104 depending on the outfit worn in the photo, with the same object valued up to 3.6 times higher …
Greyparrot raised $27 million in Series B funding on July 28, 2026, in a round led by technology investor Omar Mir. The London company plans to expand its AI waste-intelligence systems in North America and Europe, grow i…
Greyparrot.ai Ltd. raised $27 million in Series B funding led by Om Elite Assets Investment to scale its physical AI recycling infrastructure globally. The company's Greyparrot Analyzer uses AI-enabled cameras to identif…
University of Minnesota Twin Cities researchers announced July 24 an AI vision system that estimates a scuba diver's respiration rate from exhaled bubbles captured by an underwater robot. The study, published in The Inte…
Warwickshire's hospital car park now hosts the UK's first AI-enabled smart lamp-posts, called iLamps, from Conflow Power Group (CPG), which can recognize number plates and faces and are pitched as tools to fight crime an…
Snappyit launched an AI-powered product photography platform that generates ghost mannequin images for fashion and jewelry sellers, eliminating the need for studio shoots and post-production editing. The platform, recent…
Samsung revealed details about its Galaxy Smart Glasses, Android XR smart glasses launching in late 2026, which will offload heavy computing to a connected Galaxy smartphone via split computing. Head of Samsung's XR R&D …
Researchers developed the Language-guided Segmentation-assisted Diagnostic Transformer (LSDT) model, which uses zero-shot segmentation and multi-modal slice fusion to classify prostate cancer into four clinically meaning…
A new integrated deep learning and statistical framework enables whole-network gene-environment association studies of leaf vascular architecture, overcoming limitations of low-dimensional summary traits. The framework f…
Researchers propose a loss-aware feature-map pruning framework for convolutional neural networks that uses multi-armed bandits (UCB1 and Thompson Sampling) to identify and remove redundant feature maps while preserving a…
A new classification-guided large vision-language model (LVLM) framework for visual information extraction (VIE) from documents achieves 86.43% F1-score in zero-shot inference on a real-world bidding dataset with 16 cert…
Researchers introduce γ-Bridge, a look-parametric diffusion bridge for SAR despeckling that uses a schedule L(t) connecting noisy observations at L_obs to the clean limit via exact multiplicative Gamma marginals. The met…
Researchers introduce EgoPlay, an event-triggered video-to-video editor for egocentric streams that jointly learns event recognition, temporal restraint, and pixel-level editing in a single end-to-end model. On the Ego4D…
Researchers propose CausAdv, a causal framework for detecting adversarial examples in convolutional neural networks (CNNs) by learning causal and non-causal features and quantifying counterfactual information (CI) per fi…
Researchers introduce C-VCE, a diffusion framework that builds a classifier directly into the generative model via a concept bottleneck layer to produce visual counterfactual explanations without relying on external nois…
Researchers introduce FogDrive, a multi-modal synthetic driving dataset for autonomous driving under fog, containing 660 scenes (~133k frames) with four synchronized cameras, LiDAR, and radar across three calibrated fog …
Researchers propose CADER (Confidence-Aware Dynamic Evidence Reasoning), a training-free framework for adaptive long-video reasoning that uses a logit-margin signal to let high-confidence examples exit early and activate…
Researchers introduce MegaSlide-DiT, a prototype that adapts a pre-trained 105-billion-parameter Diffusion Transformer (DiT) for high-resolution video generation on a single Nvidia H200 GPU with 1.5 TB of host RAM, by ke…