Abstract:

In this work, we propose a new loss to improve feature discriminability and classification performance. Motivated by the adaptive cosine/coherence estimator [42] (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.

Links:

Citation:

J. Peeples, C. H. McCurley, S. Walker, D. Stewart and A. Zare, "Learnable Adaptive Cosine Estimator (LACE) for Image Classification," 2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2022, pp. 3757-3767, doi: 10.1109/WACV51458.2022.00381.

@InProceedings{Peeples_2022_WACV,
author    = {Peeples, Joshua and McCurley, Connor H. and Walker, Sarah and Stewart, Dylan and Zare, Alina},
title     = {Learnable Adaptive Cosine Estimator (LACE) for Image Classification},
booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
month     = {January},
year      = {2022},
pages     = {3757-3767},
doi={10.1109/WACV51458.2022.00381}
}