analiza aiqbench
This research introduces Instruct-Particulate, a neural model that significantly improves the reconstruction of articulated 3D objects by leveraging kinematic specifications to overcome data scarcity. By utilizing vision-language models for automatic part labeling and training on over 150,000 heterogeneous assets, it enables robust generalization across categories and AI-generated meshes. The advancement facilitates more realistic robotic simulations and animation pipelines where accurate joint motion prediction is critical.
Instruct-Particulate: Scaling Feed-Forward 3D Object Articulation with Kinematic Control
arXiv:2606.14699v1 Announce Type: cross Abstract: Reconstructing articulated 3D objects is important for animation, gaming, and robotic simulations. Recent neural networks can estimate the articulated structure of 3D objects, but their generalization remains limited by the scarcity of annotated data for this task. To address this gap, we introduce Instruct-Particulate, a model that takes a 3D mesh together with a target kinematic specification, including part descriptions, connectivity, joint types, and optional point prompts, and predicts the corresponding kinematic part segmentation and joint motion parameters. The kinematic specification disambiguates the task and allows the model to target annotations of different granularity, thereby making it possible to use more abundant heterogeneous training data. At test time, the kinematic specification can be obtained automatically from large-scale vision-language models, so the model can be applied to any input mesh. To train our model at scale, we construct a heterogeneous dataset of more than 150,000 articulated 3D objects, extending existing publicly available collections with data obtained by partially labelling other 3D models (monolithic or already decomposed into parts) with kinematic labels by means of vision-language models. Experiments show that our model generalizes better across categories and to AI-generated meshes, enabling articulated asset reconstruction from real-world images via image-to-3D models.
Abstract:Reconstructing articulated 3D objects is important for animation, gaming, and robotic simulations. Recent neural networks can estimate the articulated structure of 3D objects, but their generalization remains limited by the scarcity of annotated data for this task. To address this gap, we introduce Instruct-Particulate, a model that takes a 3D mesh together with a target kinematic specification, including part descriptions, connectivity, joint types, and optional point prompts, and predicts the corresponding kinematic part segmentation and joint motion parameters. The kinematic specification disambiguates the task and allows the model to target annotations of different granularity, thereby making it possible to use more abundant heterogeneous training data. At test time, the kinematic specification can be obtained automatically from large-scale vision-language models, so the model can be applied to any input mesh. To train our model at scale, we construct a heterogeneous dataset of more than 150,000 articulated 3D objects, extending existing publicly available collections with data obtained by partially labelling other 3D models (monolithic or already decomposed into parts) with kinematic labels by means of vision-language models. Experiments show that our model generalizes better across categories and to AI-generated meshes, enabling articulated asset reconstruction from real-world images via image-to-3D models.
Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR); Robotics (cs.RO)
arXiv:2606.14699 [cs.CV]
(or arXiv:2606.14699v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2606.14699 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 299 cuvinte
Subiecte conexe
Game Design Creativity evals
LessWrong · 12 oct. 2026
MolmoMotion: Language-guided 3D motion forecasting
Hugging Face · 17 iun. 2026
Latent reasoning architectures would likely undermine CoT, our strongest oversight tool
AI Alignment Forum · 23 sept. 2026
FAMOS: Feed-Forward 3D Articulation Modeling from Sparse Observations
arXiv · 18 sept. 2026