Principles of Forgetting in Domain-Incremental Semantic Segmentation in Adverse Weather Conditions

03/24/2023
by   Tobias Kalb, et al.
0

Deep neural networks for scene perception in automated vehicles achieve excellent results for the domains they were trained on. However, in real-world conditions, the domain of operation and its underlying data distribution are subject to change. Adverse weather conditions, in particular, can significantly decrease model performance when such data are not available during training.Additionally, when a model is incrementally adapted to a new domain, it suffers from catastrophic forgetting, causing a significant drop in performance on previously observed domains. Despite recent progress in reducing catastrophic forgetting, its causes and effects remain obscure. Therefore, we study how the representations of semantic segmentation models are affected during domain-incremental learning in adverse weather conditions. Our experiments and representational analyses indicate that catastrophic forgetting is primarily caused by changes to low-level features in domain-incremental learning and that learning more general features on the source domain using pre-training and image augmentations leads to efficient feature reuse in subsequent tasks, which drastically reduces catastrophic forgetting. These findings highlight the importance of methods that facilitate generalized features for effective continual learning algorithms.

READ FULL TEXT

page 6

page 8

page 13

page 14

page 15

page 16

research
09/20/2022

Improving Replay-Based Continual Semantic Segmentation with Smart Data Selection

Continual learning for Semantic Segmentation (CSS) is a rapidly emerging...
research
04/27/2023

Gradient-based Maximally Interfered Retrieval for Domain Incremental 3D Object Detection

Accurate 3D object detection in all weather conditions remains a key cha...
research
04/12/2019

ACE: Adapting to Changing Environments for Semantic Segmentation

Deep neural networks exhibit exceptional accuracy when they are trained ...
research
07/06/2020

Dynamic memory to alleviate catastrophic forgetting in continuous learning settings

In medical imaging, technical progress or changes in diagnostic procedur...
research
09/16/2022

Causes of Catastrophic Forgetting in Class-Incremental Semantic Segmentation

Class-incremental learning for semantic segmentation (CiSS) is presently...
research
02/21/2023

Effects of Architectures on Continual Semantic Segmentation

Research in the field of Continual Semantic Segmentation is mainly inves...
research
07/11/2023

MoP-CLIP: A Mixture of Prompt-Tuned CLIP Models for Domain Incremental Learning

Despite the recent progress in incremental learning, addressing catastro...

Please sign up or login with your details

Forgot password? Click here to reset