The rapid development of deep learning has made a great progress in image segmentation,
one of the fundamental tasks of computer vision. However, the current segmentation …

The existing weakly supervised semantic segmentation (WSSS) methods pay much
attention to generating accurate and complete class activation maps (CAMs) as pseudo …

Recently, over-parameterized deep networks, with increasingly more network parameters
than training samples, have dominated the performances of modern machine learning …

Unsupervised domain adaptation (UDA) approaches focus on adapting models trained on a
labeled source domain to an unlabeled target domain. In contrast to UDA, source-free …

Several factors are associated with the success of deep learning. One of the most important
reasons is the availability of large-scale datasets with clean annotations. However, obtaining …

Accurate change detection of built-up areas (BAs) fosters a comprehensive understanding of
urban development. The post-classification comparison (PCC) is a widely-used change …

Abstract Recently One-stage Weakly Supervised Semantic Segmentation (WSSS) with
image-level labels has gained increasing interest due to simplification over its cumbersome …

Due to the expensive costs of collecting labels in multi-label classification datasets, partially
annotated multi-label classification has become an emerging field in computer vision. One …

Weakly-supervised semantic segmentation aims to reduce labeling costs by training
semantic segmentation models using weak supervision, such as image-level class labels …

Existing shadow detection datasets often contain missing or mislabeled shadows, which can
hinder the performance of deep learning models trained directly on such data. To address …