Convolutional Neural Networks (CNNs) trained with the Softmax loss are widely used classification models for several vision tasks. Typically, a learnable transformation (i.e. the classifier) is placed at the end of such models returning class scores that are further normalized into probabilities by Softmax. This learnable transformation has a fundamental role in determining the network internal feature representation. In this work we show how to extract from CNNs features with the properties of maximum inter-class separability and maximum intra-class compactness by setting the parameters of the classifier transformation as not train- able (i.e. fixed). We obtain features similar to what can be obtained with the well-known OCenter LossO [1] and other similar approaches but with several practical advantages including maximal exploitation of the available feature space representation, reduction in the number of net- work parameters, no need to use other auxiliary losses besides the Softmax. Our approach unifies and generalizes into a common approach two apparently different classes of methods regarding: discriminative features, pioneered by the Center Loss [1] and fixed classifiers, firstly evaluated in [2]. Preliminary qualitative experimental results provide some insight on the potentialities of our combined strategy.
Maximally Compact and Separated Features with Regular Polytope Networks / Pernici Federico, Bruni Matteo, Baecchi Claudio, Del Bimbo Alberto. - ELETTRONICO. - (2019), pp. 46-53. (Intervento presentato al convegno Conference on Computer Vision and Pattern Recognition (CVPR) Workshops tenutosi a Long Beach, CA, USA nel 16-20 June 2019).
Maximally Compact and Separated Features with Regular Polytope Networks
Pernici Federico
;Bruni Matteo;Baecchi Claudio;Del Bimbo Alberto
2019
Abstract
Convolutional Neural Networks (CNNs) trained with the Softmax loss are widely used classification models for several vision tasks. Typically, a learnable transformation (i.e. the classifier) is placed at the end of such models returning class scores that are further normalized into probabilities by Softmax. This learnable transformation has a fundamental role in determining the network internal feature representation. In this work we show how to extract from CNNs features with the properties of maximum inter-class separability and maximum intra-class compactness by setting the parameters of the classifier transformation as not train- able (i.e. fixed). We obtain features similar to what can be obtained with the well-known OCenter LossO [1] and other similar approaches but with several practical advantages including maximal exploitation of the available feature space representation, reduction in the number of net- work parameters, no need to use other auxiliary losses besides the Softmax. Our approach unifies and generalizes into a common approach two apparently different classes of methods regarding: discriminative features, pioneered by the Center Loss [1] and fixed classifiers, firstly evaluated in [2]. Preliminary qualitative experimental results provide some insight on the potentialities of our combined strategy.File | Dimensione | Formato | |
---|---|---|---|
Pernici_Maximally_Compact_and_Separated_Features_with_Regular_Polytope_Networks_CVPRW_2019_paper.pdf
accesso aperto
Tipologia:
Pdf editoriale (Version of record)
Licenza:
Open Access
Dimensione
1.12 MB
Formato
Adobe PDF
|
1.12 MB | Adobe PDF |
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.