Anton Marini

@vade.info

ML / Video / Graphics / Complaining About Python™️ / Art / Politics / Detrash Where The Thunder Lives https://vade.info https://ozu.ai

Today fabric.graphics is having its first ever community virtual meet up at 5pm EST Curious about Fabric, or creative coding? Want to learn more? Join the discord! We’ll be discussing features, a small show and tell meeting the community. All are welcome! discord.com/invite/CrG92...

Fabric - The Creative Coding Stack for Apple Platforms

Fabric is a node-based visual programming environment and rendering engine for creative coding on Apple platforms.

fabric.graphics

On a side note, I hate everything about the opening of the new Destiny Renegades, it’s the most derivative, unoriginal, low hanging fruit I’ve ever experienced in this franchise. Good lord.

Fabric Alpha 5 is out, with preliminary @_syphon support, HDR tone mapping, color space conversion, film emulation & additional image processing effects, UI improvements and bug fixes: Fabric alpha now has over 220 nodes bundled: github.com/Fabric-Proje...

Release Releases/Alpha-5 · Fabric-Project/Fabric

New Features Preliminary Syphon support - note, may be some color matching issues Syphon Server Syphon Client Add set of Tone Mapping nodes for HDR -> SDR conversion Aces Filmic Reinhard Re...

github.com

Just released Fabric Alpha 4 which includes #MLX support, allowing use of local ML / LLMs / vLLMs optimized for Apple Silicon / GPU - we include a new local LLM node as a sample. This release also has a ton of quality of life and bug fixes, as well as a sample files: github.com/Fabric-Proje...

Release Fabric Alpha 4 · Fabric-Project/Fabric

Alpha 4 features Improved Sidebar UI Improved scrolling performance of the main canvas Improved Selection UI via Right/Control click on Nodes, allowing for Select all Nodes Upstream node Selecti...

github.com

Lately, I’ve been thinking about what’s at stake for us creative folks when the ground beneath us feels uncertain. Beginning next Wednesday, every other week, I’ll be sharing conversations with creative people about the idea of risk. Strap in!

Bild

I was reminded the other day that I’ve spent the last 11 years working on independent video ML research focused on creativity & story telling, and shipping products & experiences in that time. Mpeg7 fingerprinting, Deep belief, tensorflow, PyTorch, CoreML, vLLMs. First POC: vimeo.com/148001471

Synopsis Analysis work in progress

Synopsis is a video analysis (and optional transcoder) that runs computer vision processes and saves the results directly into a Qucktime or MP4 standards compliant…

vimeo.com

If you’ve followed me on other platforms ( 🤢 ) you might know I’ve spent the last few years doing computational cinema / edit understanding. I’m happy to say we’ve got the first public live narrative search engine for film and tv! It’s early days. But check it out! ozu.ai

Bild

Been doing more work in the necessary evils of tool building. Organizing the UI, document serialization, bug fixes and easy but boring nodes that need making. Wiring up wires.

Coffee Pro Tip: When working with raw coffee beans to grind oneself (say for a French press) one must be careful to select the right beans for the best flavor profile. Also be sure to chose coffee beans. Yes I did genuinely just reach this level of tired/auto pilot. Yes that is dog food.

Bild

A lot of ML tools help you implement. Not many help you think. When I’m exploring a new research direction, I don’t want another search engine or citation graph. I want something that’s actually read the literature, can suggest promising directions, and helps me reason through tradeoffs.