Computer Vision news and analysis on Web Pulse: 4325 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
A two-stage, training-free solution using Gemini-3.1 Pro and SAM3 ranked third in the MeViS-Text track of the 8th LSVOS Challenge, achieving J&F, J, F, N-acc., T-acc., and Final scores of 0.761, 0.7367, 0.7852, 0.8333, 0…
Researchers introduced the first agentic vision-language framework for converting structural framing plans into editable finite-element model drafts without task-specific detector training, achieving scale estimation wit…
Researchers introduced Loki-OT, a method that uses Unbalanced Optimal Transport to propagate region-level tissue reasoning to individual cell predictions, improving lymphocyte mimicry correction in pathology. On the TCGA…
Researchers propose PXDepth, a monocular depth estimation model that separates global context modeling from pixel-level prediction to preserve fine-grained structures and object boundaries. The model uses a large-patch V…
The 10th AI City Challenge, held with ECCV 2026, attracted 325 registered teams from 26 countries and regions, up from 245 teams and 15 countries in 2025, according to the challenge organizers. The benchmark suite includ…
Researchers introduced PROBE, a framework for benchmarking and finetuning vision-language model (VLM) agents on Manipulation-Grounded Visual Question Answering (MG-VQA), where robots must physically move objects to answe…
A new study from arXiv (2608.17095v1) introduces inference-time attention steering for vision-language-action (VLA) driving models, applying a bounded additive pre-softmax attention bias to visual tokens of detector-loca…
Researchers introduced UniQuery4R, a query-conditioned framework for unified 4D scene reconstruction that encodes a multi-frame clip once and selects source and target views at decoding time, achieving the best macro-ave…
Apple's macOS Tahoe 26.7 release candidate contains a video showing AirPods with built-in cameras using Visual Intelligence, according to MacRumors. The unannounced project, codenamed B790, could let users identify objec…
A developer found that their AI quality gate for locally generated images was effectively a rubber stamp, with a vision model returning identical scores across 40 images. After testing multiple models and prompt variatio…
A new interactive map, the Style Atlas, places 365 art styles on Scott McCloud's Big Triangle, using Google's Gemini 3.7 Flash multimodal model to tag images and measure how well comic language describes visual art. The …
Roboflow Inc. explains that on-premise computer vision keeps inference local, with production images processed on hardware inside the facility, which can keep sensitive data local, remove network round trips, and allow o…
Apple is preparing to launch camera-equipped AirPods, codenamed B798, that will use Visual Intelligence to provide context to Siri without recording content, according to a video clip in the macOS Tahoe 26.7 release cand…
Anker sub-brand Eufy released the Eufy Security NVR System E50, a hardwired local security camera system starting at $1,200 that includes an NVR box with at least 2TB of expandable storage (up to 16TB) and four power-ove…
Anduril Industries, a defense technology company, is hiring a Senior Computer Vision Engineer for its Frontier AI team in Washington, D.C., with a salary range of $191k–292k/yr. The role involves designing, training, and…
Xaira Therapeutics, an AI-driven biotech startup headquartered in the San Francisco Bay Area, Seattle, and London, is hiring a Robotics Engineer for Lab Automation in South San Francisco, California, with a salary of $15…
Wayve, a developer of Embodied AI technology founded in 2017, is hiring a Principal Machine Learning Engineer for Geometric Vision in Sunnyvale, offering a salary of $407k–460k/yr. The role involves building 3D foundatio…
Helsing, a defense technology company developing AI-enabled drone software, is hiring a Machine Learning Engineer for Detection and Tracking in Washington, DC, to own the detection and tracking models powering its Altra …
Clad9, a new AI wardrobe app, digitizes a user's entire closet from a single video and provides daily outfit recommendations matched to body shape, skin tone, season, weather, and calendar events. The app combines an AI …
Roboflow published a tutorial on verifying torque marks with computer vision, using its RF-DETR segmentation model and a Vision-Language Model to automate inspection. The workflow detects paint seals on fasteners and cla…