Mindy

character control, animation, and robotics

[Kaifeng Zhao @kaifengzhao]

Introducing ARDY — real-time, controllable 3D human motion generation
Streaming text prompts + flexible kinematic constraints — root waypoints, full-body keyframes, joint targets — all at interactive speed
To be presented at SIGGRAPH 2026, Paper & demo
https://t.co/ujS94CQAto
1/5

[Mindy @blossomindy]

@francoisfleuret Lots of great omakase spots! but make sure to book in advance through Catchtable app! many of the popular ones get fully booked quickly.

https://t.co/e7DwHCEOjr, https://t.co/pIW4MHOd4v, https://t.co/SFWi8C9jnh,
https://t.co/XiRJnPf7bA

[Mindy @blossomindy]

For more recommendations, check out the link below!

coffee: https://t.co/XbASrpWEOY
food: https://t.co/eoiGQr9i6A

[Mindy @blossomindy]

If you're visiting 🇰🇷 Seoul for #ICML2026, welcome!

I'm a character-control researcher and a longtime local near COEX, so I put together a Google Maps list of my favorite restaurants and cafés around the venue.

Hope it helps. Enjoy your stay in Seoul!

https://t.co/i4BQVUb6PU

[Mindy @blossomindy]

Another local tip:

If you need a break, grab a coffee and take a walk around Seonjeongneung Royal Tombs.

It’s a peaceful UNESCO World Heritage site just a short walk from COEX—perfect for a reset.

[Mindy @blossomindy]

@raila_san As a longtime local near COEX, I put together a Google Maps list of restaurants and cafés around the venue!

https://t.co/wAiCj4vE3h

[Mindy @blossomindy]

If you're visiting 🇰🇷 Seoul for #ICML2026, welcome!

I'm a character-control researcher and a longtime local near COEX, so I put together a Google Maps list of my favorite restaurants and cafés around the venue.

Hope it helps. Enjoy your stay in Seoul!

https://t.co/i4BQVUb6PU

[Mindy @blossomindy]

@taesiri Very cool! I am curious how well the agents handled the physics constraints - joints, contacts, friction, mass and inertia - when building the truck and tank.

[Mindy @blossomindy]

I’ve been following ProtoMotions from the early days, so this GPC update is really exciting!

Large-scale human motion →a discrete vocabulary of reusable motor skills → real-time control and adaptation.

Motor control is starting to look like a generative pretraining problem.

[Chen Tessler @ChenTessler]

We've released an update to ProtoMotions!
https://t.co/RASwWtyfLb

Most importantly, this release was built with Yifeng Jiang, @YiShi_333 , @erwincoumans and @xbpeng4 , whose work shaped everything from the core methods to the final code release.

1/5

[Mindy @blossomindy]

Multi-character control is harder than it looks. One of the biggest challenges is keeping characters synchronized without sacrificing physical stability.

Still a work in progress...😫 Curious how others are approaching this problem. Any recommendations?

[Ye Yuan @_ye_yuan]

Humanoid robotics is hitting a data wall. Teleop and mocap took us far, but they don’t scale to every object, terrain, and behavior.

We’re releasing GRAIL: https://t.co/LxTKtMPtw0 — a fully digital pipeline for generating loco-manipulation data before the robot moves. 🧵(1/8)

[Michael Black @Michael_J_Black]

I’ve been capturing 3D human motion for 30 years and today is maybe the biggest day in that history. We are presenting MAMMA at CVPR (oral session 2A). MAMMA is a markerless multi-camera system that has accuracy similar to marker-based systems.

[Umar Iqbal @UmarIqb]

#NVIDIA just released a whole ecosystem for human(oid) motion and robot learning from human data. 🚀🦾

Data, as we all know, is the key to scaling AI models. To accelerate the field of Embodied AI, we have open-sourced a full stack of models and tools to capture, generate, retarget, and simulate human(oid) motion data at scale, along with a massive high-quality dataset and a standard human skeletal representation, SOMA, to make them all seamlessly communicate with each other.

The entire suite is available under the Apache 2.0 license.

1️⃣ SOMA: A universal interface to unify all parametric human body models (SOMA-shape, SMPL, MHR, etc.) into a standard skeletal representation, eliminating the need for custom adapters or model-specific retargeting.
🔗 https://t.co/Xrg672T7Nu

2️⃣ Kimodo: High-fidelity, controllable text-to-motion generation for both humans and humanoid robots.
🔗 https://t.co/2cQKAPfvEU

3️⃣ GEM: A global human pose estimation method from in-the-wild videos, natively compatible with SOMA.
🔗 https://t.co/pV0043jwcO

4️⃣ Bones-SEED: A massive dataset of 150k+ motions in SOMA format, including data already retargeted for the Unitree G1, created with our partners at Bones Studio.
🔗 https://t.co/wxfyZ7S9TJ
🔗 https://t.co/oM5rIMdRi8

5️⃣ SOMA Retargeter: A dedicated tool for seamless motion retargeting from the SOMA skeleton to the Unitree G1.
🔗 https://t.co/jg4DUjWcnw

6️⃣ ProtoMotions: Our high-performance simulation framework for training digital human(oid)s via RL, now with native SOMA support.
🔗 https://t.co/K1zsGEdl5S

This is just the beginning, and we have much more in the pipeline. Excited to see what the community builds next!

#NVIDIA #GTC #GTC2026 #Robotics #EmbodiedAI #PhysicalAI @NVIDIAAI

[tingwu.wang @TingwuWang]

What is missing to bring real-time motion research into AAA games and real-world robotics?

We present MotionBricks, a step toward bridging this gap with two key components:
- a single generative latent motion backbone covering 350,000+ motion skills, running at 15,000 FPS with 2 ms latency and substantially improved quality and reliability.
- a unified smart primitive interface for locomotion, object / scene interaction, with fine-grained control over generated behaviors.

Webpage: https://t.co/aJE5skUuWD
Code: https://t.co/r56D3TJ8CW
Paper: https://t.co/CtOHXnHZMv (ACM TOG / SIGGRAPH 2026)

[Carlos DP 🤖🇺🇸 @carlosdponx]

New year robotics take #2:

For humanoid dynamics, 2025 was the year of motion tracking. We now have several methods that track target motion essentially perfectly, on robots with good sysID.

2026 will be the year of motion retargeting

In practice, current retargeting is still very flawed for motions that need to look good / survive past a demo video. You often need to manually adjust motions for the desired output, to adjust for differences in kinematics between subject and target (irrespective of the input motion quality).

In 2026, retargeting will be solved, or the goal-based methods will supersede the need to actually retarget a whole body

[Chen Tessler @ChenTessler]

Crazy we can learn to track the entire AMASS dataset in 24h.
For MaskedMimic, 1.5 years ago, it took us 2 weeks to train the tracker and another MONTH(!!!) to train the generative model.

Now it takes 24h for the tracker and another 24h for the generative model 🤯

[Chen Tessler @ChenTessler]

At @nvidia, we built ProtoMotions to help us, and researchers world-wide, innovate quickly without compromising on applicability.

We're proud to announce ProtoMotions3 -- our biggest release yet!
🧵👇

[Mindy @blossomindy]

@NliGjvJbycSeD6t Awesome! Is the paper still in progress? I'm also curious whether you combined SkillMimic with any additional policy, or explored anything beyond the base architecture

[Jason Peng @xbpeng4]

Implementing motion imitation methods involves lots of nuisances. Not many codebases get all the details right. So, we're excited to release MimicKit!
https://t.co/7enUVUkc3h
A framework with high quality implementations of our methods: DeepMimic, AMP, ASE, ADD, and more to come!

[Bernhard Schölkopf @bschoelkopf]

Dear Students, please don't get discouraged... we need you. Quoting another Geoff: "The future depends on some graduate student who is deeply suspicious of everything I have said" (8/8)