Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics Modeling
This work addresses the problem of learning rich, object-centric scene decompositions and stochastic dynamics modeling for AI systems, offering a self-supervised approach applicable to real-world multi-object datasets and decision-making.
The paper introduces Latent Particle World Model (LPWM), a self-supervised object-centric world model that learns scene decompositions like keypoints, bounding boxes, and object masks directly from video data. It models stochastic particle dynamics and achieves state-of-the-art results on various real-world and synthetic datasets, also demonstrating applicability in decision-making tasks such as goal-conditioned imitation learning.
We introduce Latent Particle World Model (LPWM), a self-supervised object-centric world model scaled to real-world multi-object datasets and applicable in decision-making. LPWM autonomously discovers keypoints, bounding boxes, and object masks directly from video data, enabling it to learn rich scene decompositions without supervision. Our architecture is trained end-to-end purely from videos and supports flexible conditioning on actions, language, and image goals. LPWM models stochastic particle dynamics via a novel latent action module and achieves state-of-the-art results on diverse real-world and synthetic datasets. Beyond stochastic video modeling, LPWM is readily applicable to decision-making, including goal-conditioned imitation learning, as we demonstrate in the paper. Code, data, pre-trained models and video rollouts are available: https://taldatech.github.io/lpwm-web