On the Road to Online Adaptation for Semantic Image Segmentation - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Chapitre D'ouvrage Année : 2022

On the Road to Online Adaptation for Semantic Image Segmentation

Résumé

We propose a new problem formulation and a corresponding evaluation framework to advance research on unsupervised domain adaptation for semantic image segmentation. The overall goal is fostering the development of adaptive learning systems that will continuously learn, without supervision, in ever-changing environments. Typical protocols that study adaptation algorithms for segmentation models are limited to few domains, adaptation happens offline, and human intervention is generally required, at least to annotate data for hyperparameter tuning. We argue that such constraints are incompatible with algorithms that can continuously adapt to different real-world situations. To address this, we propose a protocol where models need to learn online, from sequences of temporally correlated images, requiring continuous, frame-by-frame adaptation. We accompany this new protocol with a variety of baselines to tackle the proposed formulation, as well as an extensive analysis of their behaviors, which can serve as a starting point for future research.

Dates et versions

hal-04044280 , version 1 (24-03-2023)

Identifiants

Citer

Riccardo Volpi, Pau de Jorge, Diane Larlus, Gabriela Csurka. On the Road to Online Adaptation for Semantic Image Segmentation. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, pp.19162-19173, 2022, ⟨10.1109/CVPR52688.2022.01859⟩. ⟨hal-04044280⟩

Collections

MIAI ANR
15 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More