Language-Mediated, Object-Centric Representation Learning
This work provides an incremental improvement for researchers working on object-centric representation learning by leveraging language to enhance existing unsupervised object discovery methods.
This paper introduces Language-mediated, Object-centric Representation Learning (LORL), a method that integrates language input with unsupervised object discovery algorithms to learn disentangled, object-centric scene representations. LORL consistently improves the performance of unsupervised object discovery methods on two datasets and aids downstream tasks like referring expression comprehension.
We present Language-mediated, Object-centric Representation Learning (LORL), a paradigm for learning disentangled, object-centric scene representations from vision and language. LORL builds upon recent advances in unsupervised object discovery and segmentation, notably MONet and Slot Attention. While these algorithms learn an object-centric representation just by reconstructing the input image, LORL enables them to further learn to associate the learned representations to concepts, i.e., words for object categories, properties, and spatial relationships, from language input. These object-centric concepts derived from language facilitate the learning of object-centric representations. LORL can be integrated with various unsupervised object discovery algorithms that are language-agnostic. Experiments show that the integration of LORL consistently improves the performance of unsupervised object discovery methods on two datasets via the help of language. We also show that concepts learned by LORL, in conjunction with object discovery methods, aid downstream tasks such as referring expression comprehension.