45 stories tagged computer-vision, newest first.
79 points by HeavenFox | 26 comments
Landing page for AstroHelm, a mobile app that plate-solves the sky through a phone camera mounted on a telescope or camera to guide users to celestial objects.
30 points by surprisetalk | 6 comments
A blog post on bencology.bearblog.dev about building a detector to map individual and collective worm movements; the capture shows a Cloudflare verification page.
118 points by gvuksic | 28 comments
Arduino Project Hub page for JBR-001, an open-source 3D-printable desktop companion robot powered by Arduino UNO Q with camera, servos and object recognition.
355 points by plurby | 164 comments
Roboflow blog post evaluating OpenAI's GPT 5.6 Sol model on vision tasks such as object detection and OCR, calling it OpenAI's best vision model.
1608 points by verditelabs | 353 comments
Vesuvius Challenge announcement page reporting that an entire carbonized Herculaneum scroll has been virtually unrolled and read for the first time.
444 points by RIshabh235 | 178 comments
DeepSeek's chat web app login page, linked as an announcement of the chatbot gaining vision (image understanding) capabilities.
673 points by vrganj | 304 comments
DroneXL article reporting that Pokémon Go player scans helped train Niantic's visual navigation technology now being used in military drones.
802 points by ternaus | 143 comments
OpenCV team blog post announcing the release of OpenCV 5, describing its major new features for computer vision and deep learning.
382 points by HughParry | 194 comments
Anish Athalye's 2018 blog post describing Project Sistine, which turns a MacBook into a touchscreen using a $1 mirror and computer vision on the webcam.
376 points by vinhnx | 102 comments
A personal blog post explaining how JPEG image compression works by exploiting human perceptual bias through a change of basis before encoding.
325 points by SG- | 167 comments
GitHub repository for Apple's ml-sharp, a model for sharp monocular view synthesis that turns a single 2D photo into 3D views in under a second.
478 points by dvrp | 100 comments
Project page for Apple's SHARP, a method that regresses a 3D Gaussian representation from a single photo in under a second for photorealistic novel view synthesis.
522 points by thecr0w | 428 comments
Blog post describing a failed attempt to get Claude to recreate the 1996 Space Jam website from screenshots.
541 points by xnx | 277 comments
Google Keyword blog post describing Gemini 3 Pro's vision capabilities across document, spatial, screen and video understanding.
630 points by lukeinator42 | 126 comments
Meta AI's landing page introducing Segment Anything Model 3 (SAM 3), which segments and tracks objects in images and videos from text and visual prompts.
567 points by antongribok | 350 comments
Dexerto news article about armed police confronting a student after an AI weapon-detection system mistook a bag of Doritos for a gun.
508 points by zakki | 220 comments
Homepage of Frigate, an open source NVR for security cameras that performs real-time AI object detection locally on your own hardware.
330 points by marklit | 85 comments
Blog post by Mark Litwintschik examining the depth maps embedded in iPhone 15 Pro HEIC photos and extracting them with Python.
335 points by nhod | 68 comments
Apple's GitHub repository with the official implementation of FastVLM, an efficient vision encoding method for vision language models presented at CVPR 2025.
366 points by vladyslavfox | 200 comments
An interactive experiment that runs an uploaded or sample photo through Google Vision API to show how much private information a single image reveals.
349 points by robertleoj | 63 comments
A GitHub markdown write-up by a software engineer describing how he built his first robot, a vision-controlled pen plotter.
450 points by dmarcos | 131 comments
World Labs blog post introducing an AI system that generates explorable 3D worlds from a single image, as a first step toward spatial intelligence.
926 points by ulrischa | 431 comments
Simon Willison blog post marking ten years of XKCD 1425 'Tasks', noting bird photo recognition went from research-level to trivial with vision models.
674 points by xenova | 122 comments
GitHub repository from Meta's facebookresearch with inference code, model checkpoints, and example notebooks for Segment Anything Model 2 (SAM 2).
425 points by taesiri | 172 comments
Project page for the research paper "Vision language models are blind" from Auburn University and University of Alberta, with links to the arXiv paper, code, and dataset.
455 points by meetpateltech | 128 comments
Meta AI research publication page for Meta 3D Gen, a system for generating 3D assets, dated July 2, 2024, with abstract and author list.
1260 points by jimhi | 401 comments
A personal project page describing using AI object detection to drop hats from a window onto pedestrians' heads in midtown New York.
397 points by lnyan | 104 comments
Project page for Toon3D, a UC Berkeley research method for reconstructing 3D scenes from non-geometrically-consistent hand-drawn cartoon images.
319 points by nkaz123 | 78 comments
GitHub repository for Pi-C.A.R.D, a locally running AI voice assistant on a Raspberry Pi that can converse and describe camera photos.
469 points by smusamashah | 122 comments
Project page for CityGaussian, a research paper on real-time high-quality rendering of large-scale scenes using 3D Gaussian splatting, from CAS researchers.
384 points by suchintan | 134 comments
GitHub repository for Skyvern, an open-source AGPL tool that automates browser-based workflows using LLMs and computer vision.
330 points by corysama | 106 comments
Project page for a research paper from TTIC and Adobe probing whether generative image models encode intrinsic scene properties like normals, depth, albedo and shading.
1287 points by amrrs | 200 comments
Vesuvius Challenge announcement of its 2023 Grand Prize winners, who made the first Herculaneum scroll readable; the 2000-year-old text discusses music, food, and pleasure.
576 points by weirdcat | 284 comments
GitHub repository for Apple's Ferret, a multimodal large language model, containing code, scripts, and released checkpoints.
758 points by beefman | 64 comments
Project page for Visual Anagrams, a University of Michigan research paper using pretrained diffusion models to generate multi-view optical illusions.
525 points by jcparkyn | 92 comments
GitHub repository for dpoint, an open-source digital stylus that tracks six degrees of freedom using camera-based ArUco marker tracking and inertial measurements.
387 points by georgehill | 128 comments
Project page for 4K4D, a Zhejiang University research paper on real-time 4D dynamic view synthesis at 4K resolution, with paper, arXiv, code links and demos.
1095 points by razin | 204 comments
Vesuvius Challenge announcement that a 21-year-old computer science student discovered the first word read inside an unopened Herculaneum scroll.
557 points by flavoredquark | 178 comments
Project page for RealFill, a Google Research and Cornell paper on reference-driven image completion using diffusion models.
327 points by zerojames | 288 comments
Roboflow blog post giving first impressions and test results of OpenAI's GPT-4 with Vision on various computer vision tasks.
1045 points by ladino | 806 comments
OpenAI blog post announcing the rollout of new voice and image capabilities in ChatGPT, letting users talk to it and show it pictures.
353 points by squidhunter | 49 comments
A blog post describing a visual book recommender, motivated by the idea that used bookstores beat commercial outlets for discovering older, adjacent books.
813 points by jonbaer | 72 comments
Project homepage for Animated Drawings, Meta FAIR's tool and paper implementation for automatically animating children's drawings of human figures.
1265 points by kieto | 200 comments
GitHub repository for Semaphore, a Python project that uses computer vision to turn full-body gestures into keyboard typing.
651 points by crakenzak | 130 comments
Project homepage for Meta AI's Segment Anything Model (SAM), a promptable image segmentation system that can cut out any object with a single click.