Computer Vision news and analysis on Web Pulse: 4293 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
A developer built Proof of Grass, a daily outdoor bingo card web app that uses CLIP ViT-B/32 quantized to 8-bit and run in-browser via transformers.js to verify photos of outdoor items. Testing against 131 Wikimedia Comm…
A developer built TerraQuest AI, an open-source, privacy-first outdoor exploration app that uses local AI models via Ollama to help users identify nature and spend less time on screens. The app stores observations, notes…
Rob Sandhu posted on X that he gave GPT-6 Astra NFL game film and asked it to clock Marvin Harrison Jr., with yard lines and frame timestamps yielding an estimate of roughly 15 mph downfield on a touchdown stretch. Sandh…
A developer built EcoSnap, an open-source AI tool that identifies waste from a user-submitted photo and returns guidance on responsible disposal. The project runs the open-weight LLaVA 7B vision-language model locally vi…
Tavus, a Bay Area startup, said its new Griffin model fooled 48% of 54 participants in a job-interview test into thinking an AI-generated avatar was a real person, with those who believed it was human averaging 79% confi…
A developer built Rootbound: Solitary Sprouts, a 3D web app that pairs introverts with a floating gourd companion that grows when users complete solo outdoor activities. The app uses Next.js, Tailwind CSS, and Three.js f…
A developer built GrassQuest, a local-first photo scavenger hunt that uses Gemma 3 4B via Ollama to generate five nature-finding tasks, verify user-submitted photos, and write a short field journal from the confirmed fin…
Flock Safety, the AI-powered license plate reader company operating 120,000 cameras nationwide, is planning to cut roughly 18 percent of its staff, or 270 workers, according to a Reuters report. The Oakland-based advocac…
A developer's hands-on comparison of two image-to-3D tools found that Meshy AI produced a sculptural lamp mesh with 76,034 triangles and 43,446 vertices, while SupaVoxel returned roughly 1,500,000 triangles and 815,486 v…
A developer built FloraSprout, an open-source, offline-first plant care assistant that runs local Google Gemma open-weight models via LangChain and Streamlit, with a multimodal plant health scanner, garden planner, and c…
Standard Bots raised $200 million in a Series C round led by General Catalyst and RoboStrategy at a $1 billion valuation, with customers including NASA, Amazon and Lockheed Martin, co-founder and CEO Evan Beard and Head …
A developer built GrassBuddy, an open-source, offline AI plant-care assistant that runs local inference with the open-weight Llama 2 7B model and a small ResNet-based vision pipeline. The Electron desktop app diagnoses p…
A developer-authored writeup describes how AI menu analysis, using natural language processing and computer vision, can parse restaurant menu text to estimate calories, flag allergens, and rank dishes against diner prefe…
A developer built Touch Grass Quest, a daily, mobile-first web app that assigns one outdoor photo task per day and uses an open-weight vision model (Gemma, after switching from Gemini 2.5 Pro) to judge whether the submit…
A developer built PlantGuard AI, an open-source plant disease screening tool that fine-tunes a MobileNetV2 model on the PlantVillage dataset to classify 38 leaf conditions and pairs each prediction with a human-reviewed …
A developer benchmarked 13 common OCR preprocessing tricks against 10 screenshots with a known answer key, finding that none improved character error rate over the untouched original on clean images and several made resu…
A developer measured how fixed-threshold binarization silently destroys OCR accuracy, finding that a 128 cutoff erased an 11 px #999-on-#f5f5f5 footer entirely (0 characters from ImgIng's Professional, Fast and Ultimate …
A developer benchmarked 14 OCR preprocessing variants on Chinese text crops using ImgIng's Professional, Fast and Ultimate tiers on an Apple M4 with WebGPU, plus tesseract.js 5 as a control, and found that character erro…
A developer built WildLens, an offline-first progressive web app that identifies plants, birds, animals and insects from a phone photo entirely on-device. The app runs the open-weight BioCLIP vision model exported to ONN…
Tencent released Hy-Image-3.5-Preview, a unified image generation and editing model that handles text-to-image, image-to-image, and multi-turn editing through one endpoint, accepting up to 20 reference images per request…