Jae-Pil Heo* — accepted AI-conference papers

2024

Mitigating Background Shift in Class-Incremental Semantic Segmentation

Gilhan Park, WonJun Moon, SuBeen Lee, Tae-Young Kim, Jae-Pil Heo*

ECCV 2024poster

"Class-Incremental Semantic Segmentation (CISS) aims to learn new classes without forgetting the old ones, using only the labels of the new classes. To achieve this, two popular strategies are employed: 1) pseudo-labeling and knowledge distillation to preserve prior knowledge; and 2) background weig…

Cited by 2 Source ↗PDF ↗code ↗Scholar ↗

2024

Progressive Proxy Anchor Propagation for Unsupervised Semantic Segmentation

Hyun Seok Seong, WonJun Moon, SuBeen Lee, Jae-Pil Heo*

ECCV 2024poster

"The labor-intensive labeling for semantic segmentation has spurred the emergence of Unsupervised Semantic Segmentation. Recent studies utilize patch-wise contrastive learning based on features from image-level self-supervised pretrained models. However, relying solely on similarity-based supervisio…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗