Reconstruction Bottlenecks in Object-Centric Generative Models
This addresses the challenge of robust object discovery in real-world images for unsupervised learning, but it is incremental as it builds on existing models.
The study investigated how reconstruction bottlenecks affect scene decomposition in GENESIS, a VAE-based model, finding they critically influence reconstruction and segmentation quality.
A range of methods with suitable inductive biases exist to learn interpretable object-centric representations of images without supervision. However, these are largely restricted to visually simple images; robust object discovery in real-world sensory datasets remains elusive. To increase the understanding of such inductive biases, we empirically investigate the role of "reconstruction bottlenecks" for scene decomposition in GENESIS, a recent VAE-based model. We show such bottlenecks determine reconstruction and segmentation quality and critically influence model behaviour.