In this work, we propose a new loss to improve feature discriminability and classification performance. Motivated by the adaptive cosine/coherence estimator (ACE), our proposed method incorporates angular information that is inherently learned by artificial neural networks. Our learnable ACE (LACE) transforms the data into a new "whitened" space that improves the inter-class separability and intra-class compactness. We compare our LACE to alternative state-of-the art softmax-based and feature regularization approaches. Our results show that the proposed method can serve as a viable alternative to cross entropy and angular softmax approaches. Our code is publicly available: https://github.com/GatorSense/LACE.
@article{arxiv.2110.05324,
title = {Learnable Adaptive Cosine Estimator (LACE) for Image Classification},
author = {Joshua Peeples and Connor McCurley and Sarah Walker and Dylan Stewart and Alina Zare},
journal= {arXiv preprint arXiv:2110.05324},
year = {2022}
}
Comments
Accepted to WACV 2022; 14 pages (including appendix), 3 figures