LAVAE: Disentangling Location and Appearance
This work addresses the challenge of interpretable scene understanding for computer vision, representing an incremental improvement over prior disentanglement methods.
The paper tackles the problem of learning disentangled object-based representations for visual scenes, achieving full disentanglement of object location and appearance that generalizes to scenes with more objects than seen during training.
We propose a probabilistic generative model for unsupervised learning of structured, interpretable, object-based representations of visual scenes. We use amortized variational inference to train the generative model end-to-end. The learned representations of object location and appearance are fully disentangled, and objects are represented independently of each other in the latent space. Unlike previous approaches that disentangle location and appearance, ours generalizes seamlessly to scenes with many more objects than encountered in the training regime. We evaluate the proposed model on multi-MNIST and multi-dSprites data sets.