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.
Google DeepMind unveiled Gemini Robotics 2, its most advanced vision-language-action model, designed to control robots ranging from tabletop arms to full-body humanoids, along with Gemini Robotics ER 2, a higher-level re…
LocalAI's 18 custom C and C++ inference engines, including vllm.cpp and depth-anything.cpp, outperform or match upstream Python-based engines while drastically reducing footprint, with vllm.cpp delivering a 66 MiB binary…
Google Earth on the web is rolling out experimental Gemini-powered geospatial capabilities, including Ask Google Earth, a chat interface for natural-language map analysis, and Imagery search, a multimodal feature for dis…
Narcis, a free online photo tool, offers AI-powered image editing services including format conversion, compression, enhancement, and face swapping without requiring signup or watermarks. The platform provides over 25 to…
Procore Technologies Inc. will acquire DroneDeploy for approximately $845 million in cash, a deal that combines Procore's construction management software with DroneDeploy's robotics and visual intelligence platform used…
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…
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…
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…
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 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…
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…
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 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 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…
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…
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…