#computer-vision — Public Fediverse posts

Hey Geeks 👋
I just built NanoOCR for Optical Character Recognition from scratch using Kaggle OCR dataset which includes approximately ~45K PNG images and 36 classes. I use Pytorch and Colab for trained the model and used Supervised Learning. I serve the model using FastAPI. It has accuracy arround 97% which is impressive.
I want to improve it more, Please share your tips, suggestions and ideas💡 :)
Visit the Repository also 👇
Link: https://github.com/avarshvir/NanoOCR

#ComputerVision #OCR #OCRModel #AI #ML

@kaschmir @netzpolitik_feed KI Überwachung ausgehebelt - kein Fokus mehr.

Hier eine sogenannte „#Adversarial Attack“ (gegnerischer Angriff auf KI-Algorithmen). Der bunte Pullover (entwickelt von Forschern der University of Maryland) enthält ein speziell entwickeltes Muster, das Objekterkennungs-Algorithmen (wie YOLO) verwirrt. Die KI erkennt die Person dadurch nicht mehr als Mensch und löscht die Bounding Box.

#Objekterkennung #ComputerVision #Adversarialattack #KICamouflage #Datenschutz

Postdoc in AI for Medical Imaging & Simulation
Arnaout Lab @UCSF

Postdoc at the intersection of AI, modeling, and healthcare!

See the full job description on jobRxiv: https://jobrxiv.org/job/arnaout-lab-ucsf-27778-postdoc-in-ai-for-medical-imaging-simulation/

#appliedmathematics #appliedphysics #artificialintelligence #computervision #datascience #medicalimaging #ScienceJobs #hiring ...
https://jobrxiv.org/job/arnaout-lab-ucsf-27778-postdoc-in-ai-for-medical-imaging-simulation/?fsp_sid=14521

🤖🎢 Welcome to the #Roboflow Playground! Where you can have a thrilling rollercoaster ride comparing 30 computer vision models—because apparently, one isn't confusing enough. 🌀🔍 Low-code? More like no-code for understanding this ✨magical✨ AI-assisted wonderland! 🙃
https://blog.roboflow.com/roboflow-playground/ #Playground #AI #ComputerVision #NoCode #LowCode #HackerNews #ngated

Roboflow Playground: Try and Compare 30 Computer Vision Models

https://blog.roboflow.com/roboflow-playground/

Comments: https://news.ycombinator.com/item?id=49335517

#HackerNews #Roboflow #Playground #ComputerVision #AIModels #MachineLearning #TechInnovation

📢 Upcoming webinar — MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction

Guest: Jason Ren, Ai2 (Allen Institute for AI)
🗓 8/20 @ 9am Pacific
🔗 Register: https://opencv.org/molmomotion-forecasting-point-trajectories-in-3d-with-language-instruction-opencv-live-221/

Give the model an instruction in plain language and it forecasts where points travel through 3D space, building on the pointing and tracking that the Molmo family is known for.

More on Molmo: https://allenai.org/molmo

#OpenSource #MachineLearning #ComputerVision #Robotics #AI

Gemma Translator multilingual interpreter runs locally on Raspberry Pi 5 with the LiteRT runtime

https://web.brid.gy/r/https://www.cnx-software.com/2026/08/14/gemma-translator-multilingual-interpreter-runs-locally-on-raspberry-pi-5-with-the-litert-runtime/

Hey #MachineLearning / #MachineVision / #ComputerVision people.
I want to setup a pet identification system with a surveillance camera that first detects a moving object, classifies it (optionally), and uses some heuristic to specify whether the bounding box of the movement detection is one of several pets.
I don't want to pretrain a ML-model, I want to train the model guided - i.e. each time the object movement triggers, and a snapshot is created I want to assign a tag later.

And finally, I ended the season chatting to Prof. Antonia Tzemanaki (University of Bristol), Dr. George Mylonas (Imperial College London) and Prof. Tom Vercauteren (King's College London) about how robotics and AI are transforming medicine - live at the Great Exhibition Road Festival. 🤖 🩺

https://www.robottalk.org/2026/06/26/episode-162-the-robot-doctor-will-see-you-now/ #Robots #Robotics #Live #AI #Autonomy #Clinical #ComputerVision #Diagnostics #Healthcare #HRI #Rehabilitation #Surgery #Teleoperation

In June, I chatted to Dr. Maria Koskinopoulou from Heriot-Watt University about autonomous robotic manipulators for surgery, industry, and beyond. 🤖

https://www.robottalk.org/2026/06/05/episode-159-maria-koskinopoulou/ #Robots #Robotics #ComputerVision #Healthcare #Recycling #Surgery

What if a camera could process images before electronics even get involved?

Researchers have built an optical metasurface that performs key vision tasks using light itself, enabling fast, energy-efficient AI vision directly on the device.

🔗 https://www.nature.com/articles/d41586-026-01891-0

#Optics #Metasurfaces #Photonics #ComputerVision #AI

Tracking Bees with a Stereo Camera

https://fed.brid.gy/r/https://hackaday.com/2026/08/08/tracking-bees-with-a-stereo-camera/

Robotic Screw and Bolt Sorter Seeks a New Challenge

https://fed.brid.gy/r/https://hackaday.com/2026/08/08/robotic-screw-and-bolt-sorter-seeks-a-new-challenge/

Master’s, PhD, postdoc? Most postdocs would say it's a lot of hard work and discipline. According to Anna, it’s those, plus a bunch of happy accidents.

We interviewed Otto Hahn Medal winner & former @mpiinf postdoc Anna Kukleva on her journey and research.

🔗 Read more: https://sic.link/kukleva

#computervision #womeninstem #computerscience #maxplanckinstituteforinformatics #saarlandinformaticsCampus #research