Unsupervised Learning of Compositional Scene Representations from Multiple Unspecified Viewpoints
This addresses the challenge of achieving object constancy across varying viewpoints in unsupervised learning, which is incremental as it builds on existing generative models for scene understanding.
The paper tackles the problem of learning compositional scene representations from multiple unspecified viewpoints without supervision, proposing a deep generative model that separates latent representations into viewpoint-independent and viewpoint-dependent parts, and shows effectiveness on synthetic datasets.
Visual scenes are extremely rich in diversity, not only because there are infinite combinations of objects and background, but also because the observations of the same scene may vary greatly with the change of viewpoints. When observing a visual scene that contains multiple objects from multiple viewpoints, humans are able to perceive the scene in a compositional way from each viewpoint, while achieving the so-called "object constancy" across different viewpoints, even though the exact viewpoints are untold. This ability is essential for humans to identify the same object while moving and to learn from vision efficiently. It is intriguing to design models that have the similar ability. In this paper, we consider a novel problem of learning compositional scene representations from multiple unspecified viewpoints without using any supervision, and propose a deep generative model which separates latent representations into a viewpoint-independent part and a viewpoint-dependent part to solve this problem. To infer latent representations, the information contained in different viewpoints is iteratively integrated by neural networks. Experiments on several specifically designed synthetic datasets have shown that the proposed method is able to effectively learn from multiple unspecified viewpoints.