computer-vision-1

Continual Semantic Segmentation

3260 papers • 126 benchmarks • 313 datasets

Continual learning in semantic segmentation.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in continual-semantic-segmentation-1

Trend

Dataset

Best Model

Actions

ADE20K

PASCAL VOC 2012

ScanNet

Libraries

i

Use these libraries to find continual-semantic-segmentation-1 models and implementations

Datasets

ScanNet

ADE20K

Subtasks

Overlapped 5-3 Overlapped 25-25

Most implemented papers

PLOP: Learning without Forgetting for Continual Semantic Segmentation

M. Cord, Arthur Douillard, Yifu Chen, Arnaud Dapogny•Sun Nov 22 2020

Local POD is proposed, a multi-scale pooling distillation scheme that preserves long- and short-range spatial relationships at feature level that significantly outperforms state-of-the-art methods in existing CSS scenarios, as well as in newly proposed challenging benchmarks 1.

292

Content

0

Paper Graph

Tackling Catastrophic Forgetting and Background Shift in Continual Semantic Segmentation

M. Cord, Arthur Douillard, Yifu Chen, Arnaud Dapogny•Mon Jun 28 2021

Local POD is proposed, a multi-scale pooling distillation scheme that preserves long- and short-range spatial relationships at feature level that significantly outperforms state-of-the-art methods in existing CSS scenarios, as well as in newly proposed challenging benchmarks.

21 0

Paper Graph

Unsupervised Model Adaptation for Continual Semantic Segmentation

Serban Stan, Mohammad Rostami•Fri Sep 25 2020

An algorithm for adapting a semantic segmentation model that is trained using a labeled source domain to generalize well in an unlabeled target domain is developed and achieves competitive performance even compared with joint UDA approaches.

69 0

Paper Graph

SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learning

Y. Yoo, Sungmin Cha, Beomyoung Kim, Taesup Moon•Mon Jun 21 2021

This paper proposes a new method, dubbed SSUL-M (Semantic Segmentation with Unknown Label with Memory), by carefully combining techniques tailored for semantic segmentation by utilizing tiny exemplar memory for the first time in CISS to improve both plasticity and stability.

119 0

Paper Graph

Adversarial Continual Learning for Multi-Domain Hippocampal Segmentation

A. Mukhopadhyay, Marius Memmel, Camila González•Sun Jul 18 2021

This work proposes an architecture that leverages the simultaneous availability of two or more datasets to learn a disentanglement between the content and domain in an adversarial fashion, and takes inspiration from domain adaptation and combines it with continual learning for hippocampal segmentation in brain MRI.

20 0

Paper Graph

Representation Compensation Networks for Continual Semantic Segmentation

Mingg-Ming Cheng, Xialei Liu, Chang-Bin Zhang, Jianqiang Xiao, Ying-Cong Chen•Wed Mar 09 2022

This work proposes to use a structural re-parameterization mechanism, named representation compensation (RC) module, to decouple the representation learning of both old and new knowledge, and outperforms state-of-the-art performance.

127 0

Paper Graph

SATS: Self-Attention Transfer for Continual Semantic Segmentation

Yiqiao Qiu, Yixing Shen, Zhuohao Sun, Yanchong Zheng, Xiaobin Chang, Weishi Zheng, Ruixuan Wang•Mon Mar 14 2022

Extensive evaluations on multiple public benchmarks support that the proposed self-attention transfer method can further effectively alleviate the catastrophic forgetting issue, and its flexible combination with one or more widely adopted strategies significantly outperforms state-of-the-art solutions.

41 0

Paper Graph

Improving Replay-Based Continual Semantic Segmentation with Smart Data Selection

Tobias Kalb, J. Beyerer, Björn Mauthe•Mon Sep 19 2022

The findings suggest that in a class-incremental setting, it is critical to achieve a uniform distribution for the different classes in the buffer to avoid a bias towards newly learned classes, and that the effective sampling methods help to decrease the representation shift significantly in early layers, which is a major cause of forgetting in domain- increasemental learning.

14 0

Paper Graph

Label-Guided Knowledge Distillation for Continual Semantic Segmentation on 2D Images and 3D Point Clouds

K. Ma, Ruibo Li, Ze Yang, Evan Ling, Chi Zhang, Yiming Wang, Dezhao Huang, Minhoe Hur, Guosheng Lin•Sat Sep 30 2023

Continual semantic segmentation (CSS) aims to extend an existing model to tackle unseen tasks while retaining its old knowledge. Naively fine-tuning the old model on new data leads to catastrophic forgetting. A common solution is knowledge distillation (KD), where the output distribution of the new model is regularized to be similar to that of the old model. However, in CSS, this is challenging because of the background shift issue. Existing KD-based CSS methods continue to suffer from confusion between the background and novel classes since they fail to establish a reliable class correspondence for distillation. To address this issue, we propose a new label-guided knowledge distillation (LGKD) loss, where the old model output is expanded and transplanted (with the guidance of the ground truth label) to form a semantically appropriate class correspondence with the new model output. Consequently, the useful knowledge from the old model can be effectively distilled into the new model without causing confusion. We conduct extensive experiments on two prevailing CSS benchmarks, Pascal-VOC and ADE20K, where our LGKD significantly boosts the performance of three competing methods, especially on novel mIoU by up to +76%, setting new state-of-the-art. Finally, to further demonstrate its generalization ability, we introduce the first CSS benchmark for 3D point cloud based on ScanNet, along with several re-implemented baselines for comparison. Experiments show that LGKD is versatile in both 2D and 3D modalities without requiring ad hoc design. Codes are available at https://github.com/Ze-Yang/LGKD.

20 0

Paper Graph

Adding a benchmark result helps the community track progress.