Small update to #Annotorious, my #JavaScript image annotation library: You can now rotate rectangle shapes!
| Website | https://rainersimon.io |
| Github | https://github.com/rsimon |
| Website | https://rainersimon.io |
| Github | https://github.com/rsimon |
BTW: this feature was sponsored by a supporter 🙌
If you rely on open source and need something specific, consider backing it—this is how projects stay alive. Just sayin’ 😉
Small update to #Annotorious, my #JavaScript image annotation library: You can now rotate rectangle shapes!
Quick question to the Fediverse: has anyone successfully implemented image embedding in the browser *with WebGPU support*?
All my attempts seem to fail. Regardless of wether im using Transformers.js or onnxruntime-web, pre-built models or ONNX files exported myself – at runtime, processing always seems to fall back to CPU.
(Interestingly: segmentation with a self-exported FastSAM model runs fine and *does* use WebGPU.)
Gesucht: Informatiker (m/w/d)
für das Team Automatische Erschließung an der DNB
Wenn ihr jemanden kennt, if you know someone: https://stellen.dnb.de/jobposting/214c93290eb8f5c1b09fd1ea6424a0ad075a73211
Leipzig | Vollzeit | E-11 | ab sofort | unbefristet | Stellen-ID: 03/26/M 1/1464 - L

Für unser Team Automatische Erschließung suchen wir am Standort Leipzig unbefristet einen Informatiker (m/w/d). IT-Fachleute finden bei uns eine Vielfalt spannender Projekte und viel Freiraum, um innovative Lösungen für die Zukunft zu gestalten. Ihre Aufgabe besteht darin, innovative Verfahren zur automatischen Erschließung von Publikationen in der Deutschen Nationalbibliothek zu entwickeln. Mit Ihrer Expertise unterstützen Sie die Konzeption, Implementierung und Integration KI-basierter Verfahren zur automatischen Erschließung von Medienwerken. Bewerbungsfrist: 08. April 2026
RE: https://fediscience.org/@tschfflr/116124799617540823
Come work with us! You’ll be the #ComputationalLinguistics specialist (with @tschfflr, who is a great supervisor) in our interdisciplinary #Linguistics/ #DigitalHumanities/ #ReligiousStudies team.
Small update for tiny.iiif — my no-nonsense open-source IIIF server.
You can now choose your image server during setup:
• IIPImage
• Cantaloupe
(Big thanks to @atomotic for figuring out the IIPImage setup!)
For most use cases, it’s largely a matter of taste. In my experience, IIPImage tends to be slightly faster and a bit less resource-hungry, though.
Running small to mid-sized collections? Teaching with your own IIIF materials? Building IIIF-enabled tools? Check out tiny.iiif at:
It is done! 🚀 Today I’m publishing the first full, formal open source release of liiive — my web-based, real-time collaborative annotation environment for #IIIF image collections.
liiive enables users to view and annotate IIIF images together in real time — with shared cursors, synchronized drawing tools, and exportable, standards-compatible annotations.
As of today, the repository contains everything you need to run your own fully self-hosted instance:
https://github.com/rsimon/liiive
A big thank you to the University of Graz Department of Digital Humanities and the DHInfra.at initiative — this release wouldn’t have been possible without their support.
And now? it’s up to you!
Clone the repo. Deploy your own instance. Start annotating.
Most importantly: tell us how it goes. If something breaks, open an issue on GitHub. If something feels unclear, confusing, or slow — we want to know. Your feedback will shape where liiive goes next.
Yay! The first image served from my new tiny.iiif test instance.
I'm running on a 2 CPU/4GB RAM VM, which seems to be the absolute minimum, but performance is pretty slow.
If anyone has advice on what specs I'd need for getting more #IIIF speed out of Cantaloupe – let me know.