Incremental Learning of Object Detectors without Catastrophic Forgetting

Konstantin Shmelkov 1 Cordelia Schmid 1 Karteek Alahari 1
1 Thoth - Apprentissage de modèles à partir de données massives
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann
Abstract : Despite their success for object detection, convolutional neural networks are ill-equipped for incremental learning, i.e., adapting the original model trained on a set of classes to additionally detect objects of new classes, in the absence of the initial training data. They suffer from "catastrophic forgetting"---an abrupt degradation of performance on the original set of classes, when the training objective is adapted to the new classes. We present a method to address this issue, and learn object detectors incrementally, when neither the original training data nor annotations for the original classes in the new training set are available. The core of our proposed solution is a loss function to balance the interplay between predictions on the new classes and a new distillation loss which minimizes the discrepancy between responses for old classes from the original and the updated networks. This incremental learning can be performed multiple times, for a new set of classes in each step, with a moderate drop in performance compared to the baseline network trained on the ensemble of data. We present object detection results on the PASCAL VOC 2007 and COCO datasets, along with a detailed empirical analysis of the approach.
Type de document :
Communication dans un congrès
ICCV - IEEE International Conference on Computer Vision, Oct 2017, Venice, Italy. 2017
Liste complète des métadonnées


https://hal.inria.fr/hal-01573623
Contributeur : Thoth Team <>
Soumis le : jeudi 10 août 2017 - 10:46:12
Dernière modification le : vendredi 11 août 2017 - 11:34:34

Fichier

1369.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01573623, version 1

Citation

Konstantin Shmelkov, Cordelia Schmid, Karteek Alahari. Incremental Learning of Object Detectors without Catastrophic Forgetting. ICCV - IEEE International Conference on Computer Vision, Oct 2017, Venice, Italy. 2017. <hal-01573623>

Partager

Métriques

Consultations de
la notice

42

Téléchargements du document

43