Computer Vision news and analysis on Web Pulse: 1684 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
A developer describes a pipeline that processes 7,500 product images daily for eCommerce, combining AI background removal with human review for complex items. The system uses a complexity classifier to route simple produ…
RosettaSim, a new framework from researchers, integrates Large Language Model attention mechanisms into traffic simulations to improve multi-agent interactions, outperforming existing methods in the Waymo Open Sim Agent …
Researchers introduced RT-SFOD, a new object detection method that improves speed and accuracy by building on YOLOv10 and using dual-head pseudo-label fusion and multi-scale feature diversification. The approach achieves…
Researchers introduced RoMaP, a 3D Gaussian editing framework that enables precise local edits through innovative mask generation and loss techniques. The framework achieves state-of-the-art performance in local 3D editi…
Researchers have introduced diffusion crossover, a method that redefines evolutionary recombination within Denoising Diffusion Probabilistic Models (DDPMs) for generating semantically rich images. By using spherical line…
Researchers introduced Histogram-constrained Image Generation (HIG), a new method for diffusion models that uses optimal transport theory to enforce user-specified distributional constraints like color histograms with ex…
Google's Nano Banana Pro (gemini-3-pro-image-preview) AI image generation model was used to recreate two historical scenes: Julius Caesar's relief of the siege of Alesia in 52 BC and Napoleon's self-coronation in 1804. T…
Intraoperative AI tools are set to revolutionize surgical care by offering real-time insights and quality assessment, promising significant improvements in patient outcomes. The technology uses endoscopic video feeds to …
Oak Ridge National Laboratory's Advanced Plant Phenotyping Laboratory has deployed an AI framework that reduces plant image analysis from weeks to seconds, using a conversational Co-Scientist Agent and a Compute Agent ru…
Researchers propose the Quality Representation Module (QRM), a lightweight transformer module that learns a quality-aware representation to modulate diffusion transformers (DiT) for text-to-image generation, improving im…
KAIST researchers developed BehaVERT, an AI model that interprets mouse gestures by converting skeletal movements into language-like tokens, outperforming previous systems on five international benchmarks. Published in t…
GitHub trending repos this week show a shift from building new things to optimizing existing processes, with AI-driven efficiency and visual processing capabilities leading the charge. Ponytail, an ecosystem for extreme …
Google's Gemini Omni Flash API enables conversational video editing through plain language prompts, allowing users to upload footage, analyze scenes, and generate edit instructions without traditional video editing skill…
Google's Gemini Omni Flash and ByteDance's Seedance 2.5 take fundamentally different approaches to AI video, with Gemini excelling as a multimodal reasoning engine for video analysis and workflow orchestration, while See…
Orbital AI is deploying satellites equipped with artificial intelligence to monitor oceans for piracy, illegal fishing, and dumping, aiming to revolutionize maritime security. The technology detects suspicious vessels an…
Satellites equipped with onboard artificial intelligence can detect and respond to piracy, marine vandalism, and other offshore threats in real time, marking the dawn of Orbital AI. This shift from ground-based data cent…
The world's largest camera, built at SLAC National Accelerator Laboratory in Menlo Park, began capturing images for the Legacy Survey of Space and Time at the Rubin Observatory in Chile. The 10-year project will photogra…
ByteDance's Seedance V1.5 Pro Image to Video outperformed Wan v2.6 Image to Video in a head-to-head test, scoring 17.6 to 12.3. Seedance better preserved specific camera logic and prompt fidelity, while Wan often substit…
Researchers introduced GAT, a purely transformer-based Generative Adversarial Network trained in a compact Variational Autoencoder latent space, achieving state-of-the-art single-step class-conditional generation on Imag…
A developer built a Windows computer-use MCP server in pure Go, delivering a single 27 MB executable with 120 automation tools for local LLMs. The project, spanning over 14,000 lines of Go, implements vision, memory, and…