AI Art Weekly
ꜩ Collect Cover

Subscribe to our newsletter

AI Art Weekly #132

Hello, my fellow dreamers, and welcome to issue #132 of AI Art Weekly! 👋

I’ve been enjoying the hell out of Magnific’s new precision upscaler over the last two weeks for upscaling photorealistic Midjourney images. Compared to other AI upscalers, this one retains the original image extremely well, sharpens it, and adds smart noise. If you’re into photography, I definitely recommend giving it a try.

Apart from that, I don’t have any smart words for you this week. Enjoy the papers and the weekend ✌️

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

DINOv3

Meta released DINOv3 this week. A new 7B state-of-the-art computer vision model trained with self-supervised learning that enables powerful, high-resolution visual understanding capabilities like:

  • Extract detailed object masks and semantic segmentation from any image
  • Generate depth maps and monocular depth estimation
  • Track objects across video frames automatically
  • Analyze satellite imagery for environmental monitoring
  • Create dense feature maps for fine-grained image understanding
  • Perform zero-shot image classification across diverse domains

Pre-trained models and code are available on Hugging Face and GitHub.

DINOv3 example

3D

PERSONA: Personalized Whole-Body 3D Avatar with Pose-Driven Deformations from a Single Image

Arxiv, Project Page, Code

PERSONA can create personalized 3D avatars from a single image, allowing for realistic animations that reflect the subject’s identity.

PERSONA example

Matrix-3D: Omnidirectional Explorable 3D World Generation

Arxiv, Project Page, Code

Matrix-3D can generate 3D worlds from a single image or text prompt. It allows users to explore these environments in any direction and supports both quick and detailed scene creation.

Matrix-3D example

VertexRegen: Mesh Generation with Continuous Level of Detail

Arxiv, Project Page

VertexRegen can generate 3D meshes with different levels of detail by reversing the edge collapse process through vertex splitting.

VertexRegen example

Image

Voost: A Unified and Scalable Diffusion Transformer for Bidirectional Virtual Try-On and Try-Off

Arxiv, Project Page, Code (coming soon)

Voost can generate realistic images of a person wearing a target garment and remove garments from images.

Voost example

Video

ShoulderShot: Generating Over-the-Shoulder Dialogue Videos

Arxiv, Project Page

ShoulderShot can generate over-the-shoulder dialogue videos that keep characters looking the same and maintain a smooth flow between shots. It allows for longer conversations and offers more flexibility in how shots are arranged.

ShoulderShot example

Animate-X++: Universal Character Image Animation with Dynamic Backgrounds

Arxiv, Project Page, Code (coming soon)

Animate-X++ can animate characters from a single image and a pose sequence while creating dynamic backgrounds.

Animate-X++ example

Yan: Foundational Interactive Video Generation

Arxiv, Project Page

Yan can generate interactive videos in real-time at 1080p and 60fps. It allows users to edit video content through text-based interactions, making it easy to change both the structure and style of the videos.

Yan examples

RealisMotion: Decomposed Human Motion Control and Video Generation in the World Space

Arxiv, Project Page, Code (coming soon)

RealisMotion can generate human videos with realistic motions by separating four key elements: the subject, background, movement path, and actions. It uses a 3D world coordinate system for better motion editing and employs text-to-video diffusion models for high-quality results.

RealisMotion example

end of summer –c 100 –ar 62:75 –exp 34 –raw –sref Nebuloom Chromoshred GlimmerDial –profile f7afmmu –s 200; upscaled with Magnific Precision

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