Leo Bringer

ML Researcher @CraftyApesVFX 🦧 - Prev Associate Researcher @UMich 〽️ - 3D Vision, Diffusion Models, VideoGen & Character Animation 🤼‍♀️

[Leo Bringer @leo_bringer]

🚀 Our paper **MDMP** has been accepted at CVPR’25 - HuMoGen 🚀

We propose a multi-modal diffusion model that fuses textual action descriptions and 3D skeletal data to generate long-term human motion predictions, with interpretable uncertainty — paving the way for safer and smarter human-robot interaction.

Check it out!
🔗 Web: https://t.co/tKpP7RR8tX
⭐ Code: https://t.co/kKXsV2c9tU

[OpenAI @OpenAI]

Now in preview: The ChatGPT desktop app for Linux.

Use ChatGPT, ChatGPT Work, and Codex where you already work and build, with your projects and browser workflows on supported Linux systems.

[Leo Bringer @leo_bringer]

@viccpoes Congrat’s you guys, this is huge :)

[Leo Bringer @leo_bringer]

added some cool features to the https://t.co/E5duSZPL4b repo, for 3d env reconstruction and camera tracking from a single monocular video

[Zeev Farbman @ZeevFarbman]

https://t.co/iB8vUIJxvI

[Decart @DecartAI]

Introducing Lucy 2.0: a World Editing Model running at 1080p, 30FPS in realtime.

Read the technical report, API access, and try the live demo 🧵

[Leo Bringer @leo_bringer]

@levelsio @deedydas There are, just export your output as a .obj file. Or you can use blender to turn your .glb into a .obj or even .fbx

[Leo Bringer @leo_bringer]

Claude code in VSCode is just too good.

[#CVPR2026 @CVPR]

#AI research has an invisible cost: compute

Starting with #CVPR2026, authors will report their compute usage. Aggregated data will help the community understand who can participate, what is sustainable, and how resources are used, promoting more transparent & equitable research.

[Chris make some 3D scans @ChrisAtKIRI]

We just open-sourced 3DGS render blender addon v4.0 (https://t.co/1aefybjYEU)! It has been a while, but we are still updating this 3D Gaussian Splatting tool! In this version, we added a real-time rendering mode, so say goodbye to those laggy displays in the viewport! Here is an example: I imported a file that contains one million gaussians, super smooth.

[Leo Bringer @leo_bringer]

Testing WanAnimate - really impressed by the facial motion coherence and preservation of details (e.g. head shadows). Noticed some style decay in pencil sketch sequences toward the end, but it seems solvable. Exciting potential here.

[Wan @Alibaba_Wan]

1/2 Bring your characters to life like never before! We're officially launching Wan2.2-Animate, a unified model for high-fidelity character animation and replacement. The model weights and inference code are now open-source for the entire community!
• Character Animation: Feed it a character image and a reference video, and it will animate your character by precisely replicating the source's expressions and movements.
• Character Replacement: Seamlessly swap the animated character into the original video's scene. It automatically matches the lighting and color tone for perfect environmental integration.

[Vishal Thakur @vishal1thakur]

Made this cavelike restaurant 3d environment with Marble from @theworldlabs @drfeifei, just with an image

The model converts it into environments you can move around & has great object permanence.

This is done with Gaussian Splatting, more on that and results from my other prompts 👇

[Leo Bringer @leo_bringer]

Such a cool feature!! Being able to import a 3d reconstructed scene through a Mesh format in Blender is a game changer. I guess the next step is 4D dynamic scenes with deformable meshes.

[Alexandre Morgand @Almorgand]

"MILo: Mesh-In-the-Loop Gaussian Splatting for Detailed and Efficient
Surface Reconstruction"

TL;DR: differentiably extract a mesh including both vertex locations and connectivity only from Gaussian parameters. Gradient flow from the mesh to GS

[Leo Bringer @leo_bringer]

@jtydhr88 Sweet! Can you load your generated textures from models like MVAdapater?

[Leo Bringer @leo_bringer]

@AdamWHarley @Gradio Very useful!!

[Leo Bringer @leo_bringer]

@willeastcott @playcanvas Sweet!

[Leo Bringer @leo_bringer]

@jisu__nam @Khoa_NguyenTuan How do you decide which points to track? Especially when both the camera and the background is dynamic.

[Leo Bringer @leo_bringer]

@RoverXingyu Using attention layers to separate ego-motion from object-motion is a cool idea! Does it allow to stabilize static backgrounds in the 4D scenes? Do you use a different method than #DUST3R for camera pose estimation?

[Leo Bringer @leo_bringer]

@iam_chonchol looks really nice but without an open source implementation it's hard to tell how good it performs compared to the state-of-the-art

[Eustache Le Bihan @eustachelb]

Transformers support is coming today! 🤗