MuCaLe-Net: Multi categorical-Level networks to generate more discriminating features

Y. Tamaazousti (1) , Hervé Le Borgne (1) , Céline Hudelot (2)

1 DIASI (CEA, LIST) - Département Intelligence Ambiante et Systèmes Interactifs
2 MICS - Mathématiques et Informatique pour la Complexité et les Systèmes

In a transfer-learning scheme, the intermediate layers of a pre-trained CNN are employed as universal image representation to tackle many visual classification problems. The current trend to generate such representation is to learn a CNN on a large set of images labeled among the most specific categories. Such processes ignore potential relations between categories, as well as the categorical-levels used by humans to classify. In this paper, we propose Multi Categorical-Level Networks (MuCaLe-Net) that include human-categorization knowledge into the CNN learning process. A MuCaLe-Net separates generic categories from each other while it independently distinguishes specific ones. It thereby generates different features in the intermediate layers that are complementary when combined together. Advantageously, our method does not require additive data nor annotation to train the network. The extensive experiments over four publicly available benchmarks of image classification exhibit state-of-the-art performances.

Mots clés: Universal signature; Image representations; Intermediate layers; Transfer learning; Visual classification

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Y. Tamaazousti, Hervé Le Borgne, Céline Hudelot. MuCaLe-Net: Multi categorical-Level networks to generate more discriminating features. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jul 2017, Honolulu, HI, United States. pp.5282-5291, ⟨10.1109/CVPR.2017.561⟩. ⟨cea-01841669⟩