Computer Vision news and analysis on Web Pulse: 3327 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
In Roseville, California, Flock Safety's AI-powered license plate readers misread license plates in 71% of the 1,427 alerts sent to police in 2023 and 2024, according to an analysis by the Roseville Police Department. Th…
Wipe.photos, a free in-browser AI photo inpainting tool, removes unwanted objects, people, text, and logos from images using LaMa and ONNX, running entirely on the user's device via WebGPU acceleration with no signup or …
Tentomushi, an AI-powered screen recording tool, converts screen recordings into structured incident and bug reports, with automatic context capture, analysis, translation, and integration with Jira, ClickUp, Linear, and…
Icetana AI secured its first commercial sales of the Antara Core security platform, with two five-year orders from Australian partner Millennium Services Group totaling $617,000, and reported annual recurring revenue gro…
Researchers at the University of Arizona tested a convolutional neural network for low-order wavefront sensing on the Tiny Observatory for Telescope Optimization (TOTO) optical testbed, as detailed in a paper submitted t…
A new arXiv study (2607.27564v1) finds that a simple order-vote policy outperforms more complex agentic strategies for multi-image medical visual question answering, achieving 57.89% ± 0.65% final-test accuracy on MedFra…
Researchers introduced Bunraku, the first system that generates a complete Live2D character model from a single illustration, including ordered RGBA layers, deformation meshes, and keypose vertex offsets. On 50 held-out …
Researchers introduced VETO, an anti-edit cloak that disrupts joint-attention blocks in modern image-editing models like FLUX.2, and VetoBench, a benchmark evaluating defenses on both localized edits and broader contextu…
Researchers introduced CG-World, a large-scale world-state dataset and protocol derived from industrial computer graphics production pipelines, containing approximately 850,000 temporally aligned segments of 1-5 seconds …
Researchers introduced OVEarth-Bench, a benchmark for open-vocabulary Earth observation that expands category breadth and query diversity, finding that current methods perform limitedly, with MLLM-based methods achieving…
Researchers introduced 4DHumanDiff, a diffusion framework that directly generates dynamic human assets represented by 4D Gaussian Splatting (4DGS) from text prompts, avoiding video pre-generation and per-scene reconstruc…
Researchers introduced MeshFM, a feedforward framework that distills 2D features from visual foundation models into 3D, enabling rich 3D shape understanding without 3D annotations. The method, which uses a two-stage trai…
Researchers propose MedXplore, a unified framework for reliable and unbiased Generalized Category Discovery in medical imaging, achieving an average 8.5% gain in All accuracy over the strongest competing methods on multi…
ProgFormer, a hierarchical voxel-space Diffusion Transformer for longitudinal brain MRI prediction, outperforms state-of-the-art methods on ADNI, AIBL, and OASIS benchmarks by using a coarse pathway for volumetric predic…
A new arXiv preprint (2607.27566v1) finds that direct answer-only supervised fine-tuning (SFT) on MedGemma-1.5-4B is the most robust adaptation method for multi-frame medical visual question answering on the MedFrameQA b…
Researchers propose Drawing-Recode, a framework that generates Parametric CAD sequences as CAD code from raster 2D CAD drawings, outperforming existing baselines and remaining robust on scanned drawings resembling indust…
Researchers propose IGME, an efficient chained method ensemble for transferable semantic segmentation attacks, achieving competitive transferability with only one source model and faster runtime than model-ensemble attac…
Flock Safety's automated license plate recognition cameras, deployed on utility poles alongside active lanes without setback or warning signage, create a crash risk as drivers slow down, swerve, or rubberneck in response…
A Circle K customer says an AI-powered self-checkout kiosk added an $8.5 billion donation to the American Red Cross to their hot dog order, according to a Reddit post. The shopper shared a photo showing the charge alongs…
Google is expanding Gemini Omni video editing with relighting tools in Google Photos and Google Vids, allowing users to change a clip's lighting through natural-language prompts. The feature is part of a broader conversa…