Empirical generalization study: Unsupervised domain adaptation vs. domain generalization methods for semantic segmentation in the wild
For autonomous vehicles and mobile robots to safely operate in the real world, ie, the wild,
scene understanding models should perform well in the many different scenarios that can be …
scene understanding models should perform well in the many different scenarios that can be …
An In-Depth Analysis of Domain Adaptation in Computer and Robotic Vision
This review article comprehensively delves into the rapidly evolving field of domain
adaptation in computer and robotic vision. It offers a detailed technical analysis of the …
adaptation in computer and robotic vision. It offers a detailed technical analysis of the …
Exploring the Benefits of Vision Foundation Models for Unsupervised Domain Adaptation
BB Englert, FJ Piva, T Kerssies… - Proceedings of the …, 2024 - openaccess.thecvf.com
Achieving robust generalization across diverse data domains remains a significant
challenge in computer vision. This challenge is important in safety-critical applications …
challenge in computer vision. This challenge is important in safety-critical applications …
Learning to predict collision risk from simulated video data
TJ Schoonbeek, FJ Piva, HR Abdolhay… - 2022 IEEE Intelligent …, 2022 - ieeexplore.ieee.org
We propose an image-based collision risk prediction model and a training strategy that
allows training on simulated video data and successfully generalizes to real data. By doing …
allows training on simulated video data and successfully generalizes to real data. By doing …
Divide, Ensemble and Conquer: The Last Mile on Unsupervised Domain Adaptation for On-Board Semantic Segmentation
The last mile of unsupervised domain adaptation (UDA) for semantic segmentation is the
challenge of solving the syn-to-real domain gap. Recent UDA methods have progressed …
challenge of solving the syn-to-real domain gap. Recent UDA methods have progressed …
Pre-training transformers for domain adaptation
The Visual Domain Adaptation Challenge 2021 called for unsupervised domain adaptation
methods that could improve the performance of models by transferring the knowledge …
methods that could improve the performance of models by transferring the knowledge …
Hard-aware Instance Adaptive Self-training for Unsupervised Cross-domain Semantic Segmentation
The divergence between labeled training data and unlabeled testing data is a significant
challenge for recent deep learning models. Unsupervised domain adaptation (UDA) …
challenge for recent deep learning models. Unsupervised domain adaptation (UDA) …
Unsupervised Domain Adaptation for Semantic Segmentation with Global and Local Consistency
Unsupervised domain adaptation (UDA) for semantic segmentation aims to learn from
labeled synthetic data to segment the unlabeled real data. Many recent methods use …
labeled synthetic data to segment the unlabeled real data. Many recent methods use …
Training Semantic Segmentation on Heterogeneous Datasets
P Meletis, G Dubbelman - arXiv preprint arXiv:2301.07634, 2023 - arxiv.org
We explore semantic segmentation beyond the conventional, single-dataset homogeneous
training and bring forward the problem of Heterogeneous Training of Semantic …
training and bring forward the problem of Heterogeneous Training of Semantic …
[PDF][PDF] Supplementary Material–Empirical Generalization Study: Unsupervised Domain Adaptation vs. Domain Generalization Methods for Semantic Segmentation in …
FJ Piva, D de Geus, G Dubbelman - openaccess.thecvf.com
Analyzing this table, we observe that several methods are discarded because their code
was not released to the public, meaning that the only way to reproduce their results is to re …
was not released to the public, meaning that the only way to reproduce their results is to re …