Skip to main content
GameDev.net gamedev.net
Research Paper

This is an academic paper or technical research. Key findings may require technical background to fully understand.

Explore Research Radar

PRO Tired of ads? Read GameDev.net ad-free and help keep the community independent with GameDev Pro — $3/month.

arXiv cs.GR
arXiv cs.GR Research
· 6 months ago • Kaixuan Wang, Tianxing Chen, Jiawei Liu, Honghao Su, Shaolong Zhu, Minxuan Wang, Zixuan Li, Yue Chen, Huan-ang Gao, Yusen Qin, Jiawei Wang, Qixuan Zhang, Lan Xu, Jingyi Yu, Yao Mu, Ping Luo

ManiTwin: Scaling Data-Generation-Ready Digital Object Dataset to 100K

Briefing

The launch of ManiTwin, featuring 100K high-quality annotated 3D assets, marks a significant advancement in robotic manipulation capabilities. This dataset, equipped with physical properties and functional annotations, allows developers to generate diverse and simulation-ready assets efficiently.

For game developers, especially those focused on AI and robotics, this means easier access to high-quality data for training models and enhancing gameplay mechanics. The automated pipeline not only saves time but also ensures that the assets are semantically rich, paving the way for more immersive and interactive experiences.

“ManiTwin provides an efficient asset synthesis and annotation workflow.”

— Authors · Describing the benefits of the new pipeline.
What do you think?
How do you see automated asset generation impacting your workflow? What challenges have you faced in sourcing diverse data for simulations?
Start the discussion
Original source
Read on arXiv cs.GR
At a glance
what
Introduction of ManiTwin dataset with 100K 3D assets
who
Developed by Kaixuan Wang and 15 co-authors
when
Submitted on 17 Mar 2026
impact
Enhances data generation for robotic manipulation and AI
Signal Positive

The dataset significantly enhances capabilities for developers.

robotics ai 3d assets data generation
Discuss

Follow AI updates

See relevant stories in your personalized news feed.

Sign in to follow

Continue on GameDev.net

Useful next steps related to this story.

Current news From simulation to real-world deployment: Unity Simulation Pro early access Unity

Game development news without the noise

One useful weekly briefing. No daily flood.

Sending your confirmation email…

By subscribing, you agree to receive GDN Weekly by email. Confirm through the link we send. Unsubscribe anytime. See our Privacy Policy.

Discussion

Loading comments...

Also Covered By

View all 18 sources
arXiv cs.GR
arXiv cs.GR
6 months ago
arXiv cs.GR
arXiv cs.GR
6 months ago
arXiv cs.GR
arXiv cs.GR
6 months ago

Story Timeline (18 sources)

March 18, 2026
arXiv cs.GR • First
Physics-Informed Video Diffusion For Shallow Water Equations
arXiv cs.GR •
Perceptual Requirements for Low-Latency Head-Mounted Displays
arXiv cs.GR •
Toward Reliable Scientific Visualization Pipeline Construction with Structure-Aware Retrieval-Augmented LLMs
arXiv cs.GR •
Fast and Reliable Gradients for Deformables Across Frictional Contact Regimes
arXiv cs.GR •
Retrieval-Augmented Sketch-Guided 3D Building Generation
arXiv cs.GR •
TAMP-OS: An Open-Source Workflow for Tactile 3D-Printable Lithographs
arXiv cs.GR •
Improving Generative Adversarial Network Generalization for Facial Expression Synthesis
arXiv cs.GR •
Parallelised Differentiable Straightest Geodesics for 3D Meshes
arXiv cs.GR •
The Midas Touch in Gaze vs. Hand Pointing: Modality-Specific Failure Modes and Implications for XR Interfaces
arXiv cs.GR •
Volumetrically Consistent Implicit Atlas Learning via Neural Diffeomorphic Flow for Placenta MRI
arXiv cs.GR •
NanoGS: Training-Free Gaussian Splat Simplification
arXiv cs.GR •
ProgressiveAvatars: Progressive Animatable 3D Gaussian Avatars
arXiv cs.GR •
VideoMatGen: PBR Materials through Joint Generative Modeling
arXiv cs.GR •
BrickSim: A Physics-Based Simulator for Manipulating Interlocking Brick Assemblies
arXiv cs.GR • Main
ManiTwin: Scaling Data-Generation-Ready Digital Object Dataset to 100K
arXiv cs.GR •
Laplace-Beltrami Operator for Gaussian Splatting
arXiv cs.GR •
Fillerbuster: Unified Generative Scene Completion Model for Casual Captures
arXiv cs.GR •
TAUE: Training-free Noise Transplant and Cultivation Diffusion Model
Back to Feed