Learning Yourself: Class-Incremental Semantic Segmentation with Language-Inspired Bootstrapped Disentanglement

• URL: http://arxiv.org/abs/2509.00527v1
• Date: Sat, 30 Aug 2025 15:18:58 GMT
• Title: Learning Yourself: Class-Incremental Semantic Segmentation with Language-Inspired Bootstrapped Disentanglement
• Authors: Ruitao Wu, Yifan Zhao, Jia Li,
• Abstract summary: Class-Incremental Semantic (CISS) requires continuous learning of newly introduced classes while retaining knowledge of past classes.<n>This phenomenon involves Prototype-Feature Entanglement caused by semantic misalignment during the incremental process, and Background-Increment Entanglement due to dynamic data evolution.<n>We introduce a Language-inspired Bootstrapped Disentanglement framework (LBD)<n>We achieve state-of-the-art performance on both Pascal VOC and ADE20k, particularly in multi-step scenarios.
• Score: 10.912635927529218
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Class-Incremental Semantic Segmentation (CISS) requires continuous learning of newly introduced classes while retaining knowledge of past classes. By abstracting mainstream methods into two stages (visual feature extraction and prototype-feature matching), we identify a more fundamental challenge termed catastrophic semantic entanglement. This phenomenon involves Prototype-Feature Entanglement caused by semantic misalignment during the incremental process, and Background-Increment Entanglement due to dynamic data evolution. Existing techniques, which rely on visual feature learning without sufficient cues to distinguish targets, introduce significant noise and errors. To address these issues, we introduce a Language-inspired Bootstrapped Disentanglement framework (LBD). We leverage the prior class semantics of pre-trained visual-language models (e.g., CLIP) to guide the model in autonomously disentangling features through Language-guided Prototypical Disentanglement and Manifold Mutual Background Disentanglement. The former guides the disentangling of new prototypes by treating hand-crafted text features as topological templates, while the latter employs multiple learnable prototypes and mask-pooling-based supervision for background-incremental class disentanglement. By incorporating soft prompt tuning and encoder adaptation modifications, we further bridge the capability gap of CLIP between dense and sparse tasks, achieving state-of-the-art performance on both Pascal VOC and ADE20k, particularly in multi-step scenarios.

Related papers

• Towards Robust Visual Continual Learning with Multi-Prototype Supervision [24.987400887222762]
  MuproCL is a novel framework that replaces the single target with multiple, context-aware prototypes.<n>A LogSumExp aggregation mechanism allows the vision model to adaptively align with the most relevant prototype for a given image.
  arXiv  Detail & Related papers  (2025-09-19T14:24:48Z)
• Harnessing Textual Semantic Priors for Knowledge Transfer and Refinement in CLIP-Driven Continual Learning [19.210280671911278]
  Continual learning aims to equip models with the ability to learn from a stream of tasks without forgetting previous knowledge.<n>We propose a unified framework that harnesses the anti-forgetting and structured nature of textual priors to guide semantic-aware knowledge transfer.
  arXiv  Detail & Related papers  (2025-08-03T04:09:00Z)
• Unified Generative and Discriminative Training for Multi-modal Large Language Models [88.84491005030316]
  Generative training has enabled Vision-Language Models (VLMs) to tackle various complex tasks. Discriminative training, exemplified by models like CLIP, excels in zero-shot image-text classification and retrieval. This paper proposes a unified approach that integrates the strengths of both paradigms.
  arXiv  Detail & Related papers  (2024-11-01T01:51:31Z)
• ECLIPSE: Efficient Continual Learning in Panoptic Segmentation with Visual Prompt Tuning [54.68180752416519]
  Panoptic segmentation is a cutting-edge computer vision task. We introduce a novel and efficient method for continual panoptic segmentation based on Visual Prompt Tuning, dubbed ECLIPSE. Our approach involves freezing the base model parameters and fine-tuning only a small set of prompt embeddings, addressing both catastrophic forgetting and plasticity.
  arXiv  Detail & Related papers  (2024-03-29T11:31:12Z)
• Language-Driven Visual Consensus for Zero-Shot Semantic Segmentation [114.72734384299476]
  We propose a Language-Driven Visual Consensus (LDVC) approach, fostering improved alignment of semantic and visual information. We leverage class embeddings as anchors due to their discrete and abstract nature, steering vision features toward class embeddings. Our approach significantly boosts the capacity of segmentation models for unseen classes.
  arXiv  Detail & Related papers  (2024-03-13T11:23:55Z)
• Incremental Object Detection with CLIP [36.478530086163744]
  We propose a visual-language model such as CLIP to generate text feature embeddings for different class sets. We then employ super-classes to replace the unavailable novel classes in the early learning stage to simulate the incremental scenario. We incorporate the finely recognized detection boxes as pseudo-annotations into the training process, thereby further improving the detection performance.
  arXiv  Detail & Related papers  (2023-10-13T01:59:39Z)
• Expedited Training of Visual Conditioned Language Generation via Redundancy Reduction [61.16125290912494]
  $textEVL_textGen$ is a framework designed for the pre-training of visually conditioned language generation models. We show that our approach accelerates the training of vision-language models by a factor of 5 without a noticeable impact on overall performance.
  arXiv  Detail & Related papers  (2023-10-05T03:40:06Z)
• Activating the Discriminability of Novel Classes for Few-shot Segmentation [48.542627940781095]
  We propose to activate the discriminability of novel classes explicitly in both the feature encoding stage and the prediction stage for segmentation. In the prediction stage for segmentation, we learn an Self-Refined Online Foreground-Background classifier (SROFB), which is able to refine itself using the high-confidence pixels of query image.
  arXiv  Detail & Related papers  (2022-12-02T12:22:36Z)
• Modeling the Background for Incremental and Weakly-Supervised Semantic Segmentation [39.025848280224785]
  We introduce a novel incremental class learning approach for semantic segmentation. Since each training step provides annotation only for a subset of all possible classes, pixels of the background class exhibit a semantic shift. We demonstrate the effectiveness of our approach with an extensive evaluation on the Pascal-VOC, ADE20K, and Cityscapes datasets.
  arXiv  Detail & Related papers  (2022-01-31T16:33:21Z)
• Dual Prototypical Contrastive Learning for Few-shot Semantic Segmentation [55.339405417090084]
  We propose a dual prototypical contrastive learning approach tailored to the few-shot semantic segmentation (FSS) task. The main idea is to encourage the prototypes more discriminative by increasing inter-class distance while reducing intra-class distance in prototype feature space. We demonstrate that the proposed dual contrastive learning approach outperforms state-of-the-art FSS methods on PASCAL-5i and COCO-20i datasets.
  arXiv  Detail & Related papers  (2021-11-09T08:14:50Z)
• SLM: Learning a Discourse Language Representation with Sentence Unshuffling [53.42814722621715]
  We introduce Sentence-level Language Modeling, a new pre-training objective for learning a discourse language representation. We show that this feature of our model improves the performance of the original BERT by large margins.
  arXiv  Detail & Related papers  (2020-10-30T13:33:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.