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.
Eight researchers posted a preprint on arXiv on 13 August 2026 introducing Spatial Memory Agent (SMA), a framework that retrieves ranked text lessons from a memory store to improve spatial reasoning in frozen vision-lang…
Coral reef health data for 2025/26 shows mean live coral cover dropped by roughly 4% across primary monitored sectors, with high-intensity bleaching recorded in 60% of surveyed tropical zones, according to the report. Th…
The CanineWhisperer project, developed by Zenieverse, is a multimodal veterinary ethology platform that uses Gemini Vision AI to decode canine body language, acoustic analysis to translate barks and whines, and ElevenLab…
Flock Safety cofounder Garrett Langley said the company's AI-enabled license plate cameras would have solved the Nancy Guthrie case, claiming 'One hundred percent' and 'There's no doubt in my mind.' Flock Safety, which c…
A developer building fully automated explainer videos discovered that their AI-generated hero changed appearance in every scene, along with historical inaccuracies such as anachronistic ships and crowns. They solved the …
A developer built a real-time dietary analyzer using GPT-4o multimodal capabilities and OpenAI Structured Outputs, combined with FastAPI and Pydantic. The system converts smartphone photos of food into structured macronu…
The dsh-vision-router plugin, version 1.1.0, has been added to the DeepSeek Harness (dsh) plugin registry, providing text-only agents with a built-in free vision chain and pixel-level vision tools including Q&A, groundin…
The dsh-vision plugin, version 0.1.0, has been added to the DeepSeek Harness (dsh) plugin registry, providing near-native image understanding capabilities. The plugin is available via the command `dsh plugin --profile we…
A developer built an AI agent to curate dog images for training a machine learning model, automating a process that previously took weeks. The agent, created for the Google All Things Agentic Hackathon, uses Gemini to fi…
Applied Intuition, Inc., a $15 billion Silicon Valley company founded in 2017, is hiring a Research Engineer for its new grad 2027 cohort, offering $140k–200k/yr, on-site in Sunnyvale, California. The role targets recent…
A developer detailed the creation of an AI-powered document scanner and OCR pipeline for Android using Kotlin, integrating CameraX, perspective correction, image enhancement, Google ML Kit for text recognition, and AI-ba…
A developer detailed the architecture for building a real-time object detection and tracking app on Android using Kotlin and YOLO-style models. The approach separates camera capture, inference, tracking, and rendering, a…
A developer built PawWise, an AI-powered web app that analyzes photos of dogs to provide health assessments, behavior explanations, and emergency triage, aiming to address the high rates of undetected obesity and health …
A new open-world perception system for driving scenes safely handles unknown road objects by using a hierarchical taxonomy, abstraction fallback, and segmentation cross-validation, as described in a reference implementat…
Roboflow's guide to edge devices for computer vision reports that the NVIDIA Jetson Orin Nano runs Roboflow's RF-DETR at about 25 FPS for single-camera detection, while the Orin NX handles four 720p streams at 30 FPS eac…
Researchers from an undisclosed institution introduced Hybrid Domain Knowledge Fusion (HDKF), a relighting-specific training framework that fuses physics, reflectance, and realism priors from synthetic, One-Light-at-a-Ti…
Researchers propose HorizonRelight, a diffusion transformer-based framework that reframes long-horizon video relighting as temporally conditioned latent domain translation, using masked target-domain self-conditioning an…
Researchers propose V-RAE, a video representation autoencoder that builds compact generative latents on top of frozen vision foundation model representations, achieving 2.13 rFVD on K600 and outperforming all evaluated l…
Artist Simon Weckert presented Digital Camouflage in 2025, a garment using an adversarial texture to make AI person-detection systems fail to recognize its wearer. Ars Electronica describes the textile as designed to con…
Meta has filed a patent for AI smartglasses that use facial recognition to identify people and automatically generate highlight reels of events such as dinner parties, according to a patent published Thursday. The system…