SCALOR: Generative World Models with Scalable Object Representations
This work addresses the problem of modeling crowded scenes with many objects for researchers in unsupervised learning and computer vision, representing a significant advancement over prior incremental improvements.
The paper tackles the challenge of scalability in unsupervised object-oriented representation learning for videos by proposing SCALOR, a generative world model that can handle scenes with up to a hundred objects, including complex dynamic backgrounds, which is orders of magnitude more than previous state-of-the-art models.
Scalability in terms of object density in a scene is a primary challenge in unsupervised sequential object-oriented representation learning. Most of the previous models have been shown to work only on scenes with a few objects. In this paper, we propose SCALOR, a probabilistic generative world model for learning SCALable Object-oriented Representation of a video. With the proposed spatially-parallel attention and proposal-rejection mechanisms, SCALOR can deal with orders of magnitude larger numbers of objects compared to the previous state-of-the-art models. Additionally, we introduce a background module that allows SCALOR to model complex dynamic backgrounds as well as many foreground objects in the scene. We demonstrate that SCALOR can deal with crowded scenes containing up to a hundred objects while jointly modeling complex dynamic backgrounds. Importantly, SCALOR is the first unsupervised object representation model shown to work for natural scenes containing several tens of moving objects.