Incremental Learning Techniques for Semantic Segmentation (2019-07-31T00:00:00.000000Z)

TL;DR

This work formally introduces the incremental learning problem for semantic segmentation in which a pixel-wise labeling is considered and proposes various approaches working both on the output logits and on intermediate features.

Abstract

Deep learning architectures exhibit a critical drop of performance due to catastrophic forgetting when they are required to incrementally learn new tasks. Contemporary incremental learning frameworks focus on image classification and object detection while in this work we formally introduce the incremental learning problem for semantic segmentation in which a pixel-wise labeling is considered. To tackle this task we propose to distill the knowledge of the previous model to retain the information about previously learned classes, whilst updating the current model to learn the new ones. We propose various approaches working both on the output logits and on intermediate features. In opposition to some recent frameworks, we do not store any image from previously learned classes and only the last model is needed to preserve high accuracy on these classes. The experimental evaluation on the Pascal VOC2012 dataset shows the effectiveness of the proposed approaches.

Authors

Umberto Michieli

1 papers

P. Zanuttigh

1 papers

TL;DR

Abstract

Authors

References41 items

Unsupervised Domain Adaptation for Semantic Segmentation of Urban Scenes

Large Scale Incremental Learning

M2KD: Multi-model and Multi-level Knowledge Distillation for Incremental Learning

Learning Without Memorizing

Incremental Learning for Semantic Segmentation of Large-Scale Remote Sensing Data

Lifelong Learning via Progressive Distillation and Retrospection

End-to-End Incremental Learning

Incremental Training of Deep Convolutional Neural Networks

Game Theoretic Analysis of Road User Safety Scenarios Involving Autonomous Vehicles

Tree-CNN: A hierarchical Deep Convolutional Neural Network for incremental learning

Tree-CNN: A Deep Convolutional Neural Network for Lifelong Learning

Incremental Classifier Learning with Generative Adversarial Networks

Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence

Incremental Learning in Deep Convolutional Neural Networks Using Partial Network Sharing

Memory Aware Synapses: Learning what (not) to forget

Incremental Learning of Object Detectors without Catastrophic Forgetting

Gradient Episodic Memory for Continual Learning

Continual Learning with Deep Generative Replay

Overcoming catastrophic forgetting in neural networks

iCaRL: Incremental Classifier and Representation Learning

Learning without Forgetting

Active Long Term Memory Networks

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

Distilling the Knowledge in a Neural Network

Error-Driven Incremental Learning in Deep Convolutional Neural Network for Large-Scale Image Classification

Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks

Microsoft COCO: Common Objects in Context

An Empirical Investigation of Catastrophic Forgeting in Gradient-Based Neural Networks

Et al

Model compression

Learn++: an incremental learning algorithm for supervised neural networks

Catastrophic forgetting in connectionist networks

Is Learning The n-th Thing Any Easier Than Learning The First?

What is a good evaluation measure for semantic segmentation?

The PASCAL visual object classes challenge 2006 (VOC2006) results

The PASCAL Visual Object Classes Challenge

Incremental and Decremental Support Vector Machine Learning

Catastrophic Interference in Connectionist Networks: The Sequential Learning Problem

and as an in

This Paper Is Included in the Proceedings of the 12th Usenix Symposium on Operating Systems Design and Implementation (osdi '16). Tensorflow: a System for Large-scale Machine Learning Tensorflow: a System for Large-scale Machine Learning

Pre-computed weights for ResNet-101

Field of Study

Journal Information

Name

Page