Learning to Transform for Generalizable Instance-wise Invariance

arXiv:2309.16672v3 Announce Type: replace-cross
Abstract: Computer vision research has long aimed to build systems that are robust to spatial transformations found in natural data. Traditionally, this is done using data augmentation or hard-coding invariances into the architecture. However, too much or too little invariance can hurt, and the correct amount is unknown a priori and dependent on the instance. Ideally, the appropriate invariance would be learned from data and inferred at test-time.
We treat invariance as a prediction problem. Given …

abstract architecture arxiv augmentation build coding computer computer vision cs.cv cs.lg data found instance natural research robust spatial systems too little type vision vision research wise