ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression
NeurIPSOral2025
TL;DR
We revisit the texture bias hypothesis in CNNs by proposing a domain-agnostic suppression protocol, finding that contrary to prior claims, CNNs primarily rely on local shape instead of texture features.
Opening excerpt from the authors’ abstract. source
Read the paper
Topics
control