AI Art Weekly
ꜩ Collect Cover

Subscribe to our newsletter

AI Art Weekly #131

Hello, my fellow dreamers, and welcome to issue #131 of AI Art Weekly! 👋

The AI summer break is officially over! The past week has been absolutely wild with major releases from all the big labs. We got GPT-5 from OpenAI, Grok Imagine from xAI, Claude Opus 4.1 from Anthropic, ElevenLabs Music, and the jaw-dropping Genie 3 from Google DeepMind (more on that below). The research side is picking up steam too with some fascinating papers on 3D human animation and interactive tools.

Also made a new music video last week using some of these new tools :)

Time to touch grass until next week ✌️

PROMPTCACHE

Support the newsletter and unlock the full potential of AI-generated art with my curated collection of 375+ high-quality Midjourney SREF codes and 4000+ creative prompts.

promptcache.com
PROMPTCACHE

News & Papers

Highlights

Genie 3

Deepmind surprised everyone this week by showing the next iteration of their Genie world model series: Genie 3. The model is now capable of generating consistent worlds that can hold their state for one minute and are interactable in real time. I knew sooner or later that we would get there, but I didn’t suspect so soon. What a few months ago was still scifi has suddendly become reality, crazy.

Painting a wall and keeping it consistent

3D

Event-Driven Storytelling

Arxiv, Project Page, Code

Event-Driven Storytelling can generate realistic movements for multiple characters in a 3D scene. It uses a large language model to understand complex interactions, allowing for diverse and scalable behavior planning based on character relationships and their positions.

Event-Driven Storytelling with Multiple Lifelike Humans in a 3D Scene example

DPoser-X: Diffusion Model as Robust 3D Whole-body Human Pose Prior

Arxiv, Project Page, Code

DPoser-X can generate and complete 3D whole-body human poses using a diffusion-based model.

DPoser-X example

GestureHYDRA: Semantic Co-speech Gesture Synthesis via Hybrid Modality Diffusion Transformer and Cascaded-Synchronized Retrieval-Augmented Generation

Arxiv, Project Page

GestureHYDRA can generate clear hand gestures for speech applications.

GestureHYDRA example

BANG: Dividing 3D Assets via Generative Exploded Dynamics

Arxiv, Project Page

BANG can break down 3D objects into parts using a method called Generative Exploded Dynamics, which keeps the shapes and meanings clear.

BANG example

MemoryTalker: Personalized Speech-Driven 3D Facial Animation via Audio-Guided Stylization

Arxiv, Project Page, Code (coming soon)

MemoryTalker can generate realistic 3D facial animations from audio alone, without needing speaker ID or 3D facial meshes.

MemoryTalker example

HairCUP: Hair Compositional Universal Prior for 3D Gaussian Avatars

Arxiv, Project Page

HairCUP can generate 3D head avatars with detailed hair and face features. It allows easy swapping of hair and face parts between avatars while keeping their identity.

HairCUP example

PINO: Person-Interaction Noise Optimization for Long-Duration and Customizable Motion Generation of Arbitrary-Sized Groups

Arxiv, Project Page, Code (coming soon)

PINO can generate realistic interactions among groups of any size by breaking down complex actions into simple pairwise motions. It uses pretrained diffusion models for two-person interactions and ensures realistic movement with physics-based rules, allowing control over character speed and position.

PINO example

Image

Qwen-Image

Arxiv, Code, Demo, Model

Qwen-Image can generate high-quality images and edit them in advanced ways. It can transfer styles, manipulate objects, and edit text in images, while also handling complex text rendering in multiple languages.

Qwen-Image Technical Report example

SDMatte: Grafting Diffusion Models for Interactive Matting

Arxiv, Code

SDMatte can extract objects from images using visual prompts like points, boxes, and masks.

SDMatte example

Video

VideoColorGrading

Arxiv, Code

VideoColorGrading can generate a look-up table (LUT) for matching colors between reference scenes and input videos.

Video Color Grading via Look-Up Table Generation example

Enjoy your weekend my fellow dreamers! Sunwrinkle style on Promptcache.

And that my fellow dreamers, concludes yet another AI Art weekly issue. If you like what I do, you can support me by:

  • Sharing it 🙏❤️
  • Following me on Twitter: @dreamingtulpa
  • Buying me a coffee (I could seriously use it, putting these issues together takes me 8-12 hours every Friday 😅)
  • Buying my Midjourney prompt collection on PROMPTCACHE 🚀
  • Buying access to AI Art Weekly Premium 👑

Reply to this email if you have any feedback or ideas for this newsletter.

Thanks for reading and talk to you next week!

– dreamingtulpa

by @dreamingtulpa