CVAILGOct 12, 2022

SlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Models

Georgia TechNVIDIAU of Toronto
arXiv:2210.05861v2148 citationsh-index: 58
Originality Incremental advance
AI Analysis

This addresses the challenge of understanding and simulating complex object interactions in videos for applications in AI and robotics, representing an incremental advancement in object-centric dynamics modeling.

The authors tackled the problem of modeling object dynamics from visual observations by introducing SlotFormer, a Transformer-based autoregressive model that predicts future object states from learned object-centric representations, achieving significantly better long-term synthesis and improving performance on downstream tasks like Visual Question Answering and planning.

Understanding dynamics from visual observations is a challenging problem that requires disentangling individual objects from the scene and learning their interactions. While recent object-centric models can successfully decompose a scene into objects, modeling their dynamics effectively still remains a challenge. We address this problem by introducing SlotFormer -- a Transformer-based autoregressive model operating on learned object-centric representations. Given a video clip, our approach reasons over object features to model spatio-temporal relationships and predicts accurate future object states. In this paper, we successfully apply SlotFormer to perform video prediction on datasets with complex object interactions. Moreover, the unsupervised SlotFormer's dynamics model can be used to improve the performance on supervised downstream tasks, such as Visual Question Answering (VQA), and goal-conditioned planning. Compared to past works on dynamics modeling, our method achieves significantly better long-term synthesis of object dynamics, while retaining high quality visual generation. Besides, SlotFormer enables VQA models to reason about the future without object-level labels, even outperforming counterparts that use ground-truth annotations. Finally, we show its ability to serve as a world model for model-based planning, which is competitive with methods designed specifically for such tasks.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes