Learning Online Visual Invariances for Novel Objects via Supervised and Self-Supervised Training
This addresses the problem of enabling AI vision systems to generalize like humans for object recognition with limited data, though it is incremental as it builds on existing CNN and invariance research.
The paper investigates whether standard CNNs can achieve human-like online invariance for novel objects after minimal training, showing that supervised CNNs trained on as few as 50 objects from 10 classes acquire strong invariances across transformations like rotation and scaling, which extends to real-world datasets, and self-supervised methods also achieve this.
Humans can identify objects following various spatial transformations such as scale and viewpoint. This extends to novel objects, after a single presentation at a single pose, sometimes referred to as online invariance. CNNs have been proposed as a compelling model of human vision, but their ability to identify objects across transformations is typically tested on held-out samples of trained categories after extensive data augmentation. This paper assesses whether standard CNNs can support human-like online invariance by training models to recognize images of synthetic 3D objects that undergo several transformations: rotation, scaling, translation, brightness, contrast, and viewpoint. Through the analysis of models' internal representations, we show that standard supervised CNNs trained on transformed objects can acquire strong invariances on novel classes even when trained with as few as 50 objects taken from 10 classes. This extended to a different dataset of photographs of real objects. We also show that these invariances can be acquired in a self-supervised way, through solving the same/different task. We suggest that this latter approach may be similar to how humans acquire invariances.