Spatially Consistent Representation Learning

Understand

Self-supervised learning has been widely used to obtain transferrable representations from unlabeled images.

• Especially, recent contrastive learning methods have shown impressive performances on downstream image classification tasks.
• While these contrastive methods mainly focus on generating invariant global representations at the image-level under semantic-preserving transformations, they are prone to overlook spatial consistency of local representations and therefore have a limitation in pretraining for localization tasks such as object detection and instance segmentation.
• Moreover, aggressively cropped views used in existing contrastive methods can minimize representation distances between the semantically different regions of a single image.