Evolving Knowledge Mining for Class Incremental Segmentation
Not on the map. Why?
Abstract (arXiv)
Class Incremental Semantic Segmentation (CISS) has been a trend recently due to its great significance in real-world applications. Although the existing CISS methods demonstrate remarkable performance, they either leverage the high-level knowledge (feature) only while neglecting the rich and diverse knowledge in the low-level features, leading to poor old knowledge preservation and weak new knowledge exploration; or use multi-level features for knowledge distillation by retraining a heavy backbone, which is computationally intensive. In this paper, we for the first time investigate the efficient multi-grained knowledge reuse for CISS, and propose a novel method, Evolving kNowleDge minING (ENDING), employing a frozen backbone. ENDING incorporates two key modules: evolving fusion and semantic enhancement, for dynamic and comprehensive exploration of multi-grained knowledge. Evolving fusion is tailored to extract knowledge from individual low-level feature using a personalized lightweight network, which is generated from a meta-net, taking the high-level feature as input. This design enables the evolution of knowledge mining and fusing when applied to incremental new classes. In contrast, semantic enhancement is specifically crafted to aggregate prototype-based semantics from multi-level features, contributing to an enhanced representation. We evaluate our method on two widely used benchmarks and consistently demonstrate new state-of-the-art performance. The code is available at https://github.com/zhiheLu/ENDING_ISS.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 55 of its 68 references that are in the map. First posted on arXiv 3 Jun 2023.
Builds on
55 of 68 references.
References · 55
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
- 2014Microsoft COCO: Common Objects in Contextcited by 17,498
- 2015Distilling the Knowledge in a Neural Networkcited by 10,076
- 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
- 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
- 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
- 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
- 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
- 2016Pyramid Scene Parsing Networkcited by 2,472
- 2016Learning without Forgettingcited by 2,465
- 2021Masked-attention Mask Transformer for Universal Image Segmentationcited by 2,338
38 more
- 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,313
- 2017Non-local Neural Networkscited by 2,290
- 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,154
- 2017Gradient Episodic Memory for Continual Learningcited by 1,871
- 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,417
- 2020Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformerscited by 1,368
- 2017Continual Learning with Deep Generative Replaycited by 1,152
- 2021Per-Pixel Classification is Not All You Need for Semantic Segmentationcited by 1,148
- 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,128
- 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,094
- 2018Experience Replay for Continual Learningcited by 1,074
- 2018Dual Attention Network for Scene Segmentationcited by 1,001
- 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 882
- 2021Segmenter: Transformer for Semantic Segmentationcited by 855
- 2018Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigencecited by 734
- 2018CCNet: Criss-Cross Attention for Semantic Segmentationcited by 717
- 2018Overcoming catastrophic forgetting with hard attention to the taskcited by 672
- 2016Expert Gate: Lifelong Learning with a Network of Expertscited by 369
- 2021Rainbow Memory: Continual Learning with a Memory of Diverse Samplescited by 273
- 2018Learning without Memorizingcited by 249
- 2020Semantic Drift Compensation for Class-Incremental Learningcited by 241
- 2016Deeply supervised salient object detection with short connectionscited by 196
- 2020Modeling the Background for Incremental Learning in Semantic Segmentationcited by 189
- 2020PLOP: Learning without Forgetting for Continual Semantic Segmentationcited by 177
- 2020Using Hindsight to Anchor Past Knowledge in Continual Learningcited by 160
- 2020Continual Prototype Evolution: Learning Online from Non-Stationary Data Streamscited by 143
- 2019Incremental Learning Techniques for Semantic Segmentationcited by 141
- 2014Hypercolumns for Object Segmentation and Fine-grained Localizationcited by 128
- 2017Incremental Learning Through Deep Adaptationcited by 108
- 2020Memory-Efficient Incremental Learning Through Feature Adaptationcited by 107
- 2020Continual Deep Learning by Functional Regularisation of Memorable Pastcited by 104
- 2017Encoder Based Lifelong Learningcited by 102
- 2021Continual Semantic Segmentation via Repulsion-Attraction of Sparse and Disentangled Latent Representationscited by 96
- 2021On Learning the Geodesic Path for Incremental Learningcited by 85
- 2021RECALL: Replay-based Continual Learning in Semantic Segmentationcited by 83
- 2022Representation Compensation Networks for Continual Semantic Segmentationcited by 72
- 2021SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learningcited by 71
- 2022Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentationcited by 28
50 most cited of 55 All 55 (JSON)
Built on by
Not cited yet.