JustCaptcha

Full-Stack Developer
#Rust #TS #JS

[Erik @sweriko]

no animations or shaders,
just a single 8bit image and color cycling.
made for 30+ year old hardware

[cocktail peanut @cocktailpeanut]

FramePack: Generate Video Forever

[NVIDIA ONLY] The creator of ControlNet just dropped a new approach to generating videos in a PROGRESSIVE way--you can generate loooong videos with LOW VRAM (just 6GB)

@SUP3RMASS1VE wrote a 1-click launcher to run the Gradio app on your PC!

[aleha_84 @84Aleha]

Path #pixelart

What or who is waiting for you in the house on top of the hill?

[cocktail peanut @cocktailpeanut]

Fluxgym Update!

1. Train on top of ANY Flux model: dev, schnell, dev2pro, and more!
2. Auto-download models from Huggingface, now built into the app.
3. 8GB VRAM Possible?

[Xenova @xenovacom]

Introducing Whisper Timestamped: Multilingual speech recognition with word-level timestamps, running 100% locally in your browser thanks to 🤗 Transformers.js!

This unlocks a world of possibilities for in-browser video editing! 🤯 What will you build? 😍

Demo (+ source code) 👇

[Simo Ryu @cloneofsimo]

Reminder, my friends.

Tomorrow you will be gifted with SD3, 2B ver.
However,

It will be tiny.

It will be closed source

It will require "license".

.
.
.

Mine won't.

Mine will be LARGE AF.

training code? its ALREADY OPEN.

mine will be MIT at best, AL2 at worst.

[JustCaptcha @JustCaptcha]

@CosmoMidias @EzzellVFX Hi Cosmo Mídias! If I'm not late, please DM me. :)

[Bronze Giant @RjNol]

Healing through food.
Is this the connection to strengthen our organism?

[Ethan @torchcompiled]

Created a new method of generative model (although kinda crappy lol) that works by autoregressive sequencing fourier coefficients. Inspired by the coarse to fine generation by Diffusion models.

Full write up here: https://t.co/voD78qVwIB

and tldr in thread 🧵

(flowers102)

[Zengyi Qin @qinzytech]

Super quality Text-to-Speech library #MeloTTS is open-sourced. Multi-lingual, multi-accent, CPU-realtime, and completely free.

It supports English, Spanish, French, Chinese, Japanese and Korean. Take a trip through seminal moments in open source history that led to Linux, narrated by #MeloTTS.

Official demo: https://t.co/a86vVyGE8r

HF space by @realmrfakename: https://t.co/IUMIpJzLXB

Github repo: https://t.co/1mT68SSh5F

A heartfelt thank you to @ForrestKnight for crafting this breathtaking documentary of Linux.

[cocktail peanut @cocktailpeanut]

Video Background Removal 1 Click Launcher

Just ported this huggingface gradio app to run locally (Win, Mac, Linux)

Also, got rid of the 1 minute limit from the original app. Run locally with this 1 click launcher to remove backgrounds for videos with NO LIMIT, as long as you want!

[A.I.Warper @AIWarper]

Kijai has implemented another method for extracting Depth / Normals through "Intrinsic Lora"

The Normal map extraction is of particular interest as it is significantly more impressive that the existing preproc

Link 👇

[brett goldstein @thatguybg]

Sora's video quality seems impossible so I dug into how it works under the hood

it uses both diffusion (starting with noise, refining towards a desired video) and transformer architectures (handling sequential video frames)

read on 🧵

[Junior Rojas @junior_rojas_d]

visualizing neuromuscular activations

[JustCaptcha @JustCaptcha]

@nolnad Happy Birthday fellow Capricorn!

[CiaraRowles @CiaraRowles1]

Announcing Stable Video Diffusion Temporal ControlNet!
You can now give stable video diffusion clips custom motion using a depth map input!
Inference code and initial weights available!
https://t.co/8rzTPXzCdo
#stablediffusion #SVD

[bolek_3d @TomaszBolek]

iron cage

[Santiago @svpino]

Like I said. Embarrassing.

[Alex Carlier @alexcarliera]

PatchFusion was just released. Compared to ZoeDepth, it predicts depth maps with much finer details, just look at the comparison below! 🤯

Its main contribution is a new Global-to-Local module and Consistency-Aware Training.

More examples below (with code) ⬇️⬇️