SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.14115 · CVPR · 2023

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

Jürgen Beyerer, Tobias Kalb

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

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.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2303.14115")
get_code_for_paper("2303.14115")
have("2303.14115")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2303.14115.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2303.14115.svg)](https://syntology.ai/paper/2303.14115)

Connect an agent — have() is free.