Computer Vision news and analysis on Web Pulse: 3344 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Adobe is rolling out a limited test of its AI Playground, an experimental set of generative AI photo-editing tools, to select users of its Project Indigo camera app. The free feature, available for a few weeks to a few p…
Adobe's experimental Project Indigo camera app, originally designed to provide a natural SLR-like look for iPhone photography, has been updated with generative AI tools in version 1.1, including filter effects, object re…
NVIDIA released Cosmos 3 Edge, a 4-billion-parameter open world model on Hugging Face, designed to deliver data center-level performance on memory-constrained edge devices for robots and vision AI agents. The model achie…
Adobe is adding AI-powered features to its experimental iOS camera app Project Indigo, including photo critiques and editing suggestions using large language models, advanced object removal, depth of field generation, an…
NVIDIA has introduced Omniverse libraries, a set of software components that bring physical AI capabilities into existing 3D applications and prepare content for simulation, targeting developers and engineers in robotics…
RoboFlow recommends profiling the full AI inference pipeline before optimizing, as the biggest savings in video applications often come from lowering frame rates with tracking, routing uncertain cases to large models, ca…
Modern browsers can now run numerical computing, machine learning, and GPU-accelerated workloads locally, eliminating the need for server infrastructure and network requests. At HPSFCon 2026, Gunj Joshi of stdlib demonst…
At SIGGRAPH, NVIDIA announced advances in graphics and simulation with agentic and physical AI, including Model Context Protocol connections that let AI agents work inside creative tools, a new Synthetic Video Detector N…
NVIDIA has released ovrtx, an RTX sensor simulation library now available as prerelease software on GitHub, enabling developers to generate camera, lidar, radar, and related sensor outputs from OpenUSD content within exi…
World Vision Zambia MERL manager Chanda Mwila won funding to develop a multi-modal AI-enabled app for Community Health Workers in rural Zambia that uses NLP and computer vision for diagnosis and triage across health, nut…
Baseus is offering its X1 Pro Dual-Tracking Wireless Solar Outdoor Security Camera for $129.99 shipped via Amazon, down from a $270 full price and just $10 above the Prime Day low of $120. The camera features dual indepe…
PaddlePaddle requests adding the Real5-OmniDocBench framework and benchmark entry to Hugging Face's Hub Benchmark evaluation system. The benchmark evaluates document parsing robustness under five real-world acquisition s…
AI startup Decart released Lucy 2.5, a live video model that generates real-time visual effects such as fireballs and smoke, and also improves physics simulations for robotics training. The model offers finer control ove…
Flock Safety's license plate recognition cameras wrongly flagged a journalist's vehicle as stolen due to a partial plate match, leading to his arrest, and the company's CEO apologized for calling privacy advocates terror…
Vancouver-based personal training app Flex AI is shutting down after failing to secure an acquisition and exhausting its funding, co-founder and CEO Amin Niri told employees in a March 24 email. The AI-powered fitness te…
Google's DeepMind lab has used artificial intelligence to recreate a goal by Brazilian soccer legend Pelé that he called his greatest ever but was never filmed.
A tiny AI vision camera based on the STM32F405RGT6 microcontroller has been developed, enabling edge AI computer vision on low-power devices. The camera integrates a camera module and runs neural network inference locall…
ImprovedVBGS, an accelerated framework for on-the-fly continual reconstruction, reduces mean per-frame latency from ~84.0 s to ~0.050 s on an RTX 3070 Ti using the NeRF synthetic dataset, a 1680x speed-up while maintaini…
Researchers propose E3DGS, a unified geometric-photometric equivariant architecture for 3D Gaussian Splatting that treats color as a geometric entity via a matrix embedding, achieving SE(3) equivariance without Clebsch-G…
Researchers propose PhysFlood, a physics-aware masked diffusion-based simulation system that generates realistic flood images from a single fisheye lens photo, addressing the shortage of flood data for urban disaster det…