Abstract
Understanding which inductive biases could be useful for the unsupervised learning of object-centric representations of natural scenes is challenging. Here, we use neural style transfer to generate datasets where objects have complex textures while still retaining ground-truth annotations. We find that, when a model effectively balances the importance of shape and appearance in the training objective, it can achieve better separation of the objects and learn more useful object representations.
Abstract (translated)
URL
https://arxiv.org/abs/2204.08479