Unsupervised Object-Based Transition Models for 3D Partially Observable Environments
This addresses the challenge of object identity and persistence in partially observable 3D environments for robotics and AI applications, representing an incremental improvement over existing methods.
The paper tackles the problem of modeling object transitions in 3D partially observable environments by introducing an unsupervised, object-based model that aligns objects to maintain identity and persistence. It outperforms a state-of-the-art baseline, effectively handling object occlusion and re-appearance.
We present a slot-wise, object-based transition model that decomposes a scene into objects, aligns them (with respect to a slot-wise object memory) to maintain a consistent order across time, and predicts how those objects evolve over successive frames. The model is trained end-to-end without supervision using losses at the level of the object-structured representation rather than pixels. Thanks to its alignment module, the model deals properly with two issues that are not handled satisfactorily by other transition models, namely object persistence and object identity. We show that the combination of an object-level loss and correct object alignment over time enables the model to outperform a state-of-the-art baseline, and allows it to deal well with object occlusion and re-appearance in partially observable environments.