Resolving Long-Tail Ambiguity in Unsupervised 3D Point Cloud Segmentation with Language Priors

• URL: http://arxiv.org/abs/2605.20737v1
• Date: Wed, 20 May 2026 05:42:29 GMT
• Title: Resolving Long-Tail Ambiguity in Unsupervised 3D Point Cloud Segmentation with Language Priors
• Abstract summary: LangTail is a language-guided hierarchical learning framework.<n>It mitigates long-tail ambiguity in unsupervised 3D segmentation.<n>It consistently outperforms existing methods by significant margins.
• Score: 21.92314835722451
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Existing approaches for unsupervised 3D point cloud segmentation predominantly rely on a purely visual similarity-based learning-by-clustering paradigm, which suffers from a fundamental limitation: long-tail ambiguity. In such a paradigm, features of minor classes are consistently absorbed by dominant clusters, leading to severely imbalanced predictions. To address this issue, we propose LangTail, a language-guided hierarchical learning framework that leverages the balanced world knowledge encoded in language models to mitigate long-tail ambiguity in unsupervised 3D segmentation. The key idea is to establish multi-level associations between language-derived semantic priors and visually underrepresented minor classes, thereby compensating for the biased attention of purely visual clustering toward dominant classes. Specifically, LangTail first constructs an entity-level semantic prior from language models, capturing balanced and fine-grained world knowledge across categories. These priors are injected into a hierarchical clustering framework via contrastive alignment. This guides multi-granularity semantic structure formation and prevents minor classes from being absorbed by dominant clusters, yielding more discriminative representations for underrepresented categories. Extensive experiments on ScanNet-v2, S3DIS, and nuScenes demonstrate that LangTail consistently outperforms existing methods by significant margins, \ie, +13.5, +12.9, and +8.9 mIoU, respectively. These results demonstrate the effectiveness of language priors in improving the representation of minority classes in 3D point clouds. The code will be released at: https://github.com/Whisky0129/langtail_official.

Related papers

• Class Distillation with Mahalanobis Contrast: An Efficient Training Paradigm for Pragmatic Language Understanding Tasks [1.1060425537315088]
  We propose textbfClass textbfDistillation (ClaD), a novel training paradigm that targets the core challenge: distilling a small, well-defined target class.<n>ClaD integrates two key innovations: (i) a loss function informed by the structural properties of class distributions, based on Mahalanobis distance, and (ii) an interpretable decision algorithm optimized for class separation.
  arXiv  Detail & Related papers  (2025-05-17T04:34:19Z)
• SemiDAViL: Semi-supervised Domain Adaptation with Vision-Language Guidance for Semantic Segmentation [9.311853182451289]
  We propose a language-guided Semi-supervised Domain Adaptation (SSDA) setting for semantic segmentation.<n>We harness the semantic generalization capabilities inherent in vision-language models (VLMs) to establish a synergistic framework.<n>Our approach demonstrates substantial performance improvements over contemporary state-of-the-art (SoTA) methodologies.
  arXiv  Detail & Related papers  (2025-04-08T19:14:34Z)
• Segment Any 3D Object with Language [58.471327490684295]
  We introduce Segment any 3D Object with LanguagE (SOLE), a semantic geometric and-aware visual-language learning framework with strong generalizability. Specifically, we propose a multimodal fusion network to incorporate multimodal semantics in both backbone and decoder. Our SOLE outperforms previous methods by a large margin on ScanNetv2, ScanNet200, and Replica benchmarks.
  arXiv  Detail & Related papers  (2024-04-02T17:59:10Z)
• Class-Imbalanced Semi-Supervised Learning for Large-Scale Point Cloud Semantic Segmentation via Decoupling Optimization [64.36097398869774]
  Semi-supervised learning (SSL) has been an active research topic for large-scale 3D scene understanding. The existing SSL-based methods suffer from severe training bias due to class imbalance and long-tail distributions of the point cloud data. We introduce a new decoupling optimization framework, which disentangles feature representation learning and classifier in an alternative optimization manner to shift the bias decision boundary effectively.
  arXiv  Detail & Related papers  (2024-01-13T04:16:40Z)
• Generalized Robot 3D Vision-Language Model with Fast Rendering and Pre-Training Vision-Language Alignment [55.11291053011696]
  This work presents a framework for dealing with 3D scene understanding when the labeled scenes are quite limited.<n>To extract knowledge for novel categories from the pre-trained vision-language models, we propose a hierarchical feature-aligned pre-training and knowledge distillation strategy.<n>In the limited reconstruction case, our proposed approach, termed WS3D++, ranks 1st on the large-scale ScanNet benchmark.
  arXiv  Detail & Related papers  (2023-12-01T15:47:04Z)
• Towards Realistic Zero-Shot Classification via Self Structural Semantic Alignment [53.2701026843921]
  Large-scale pre-trained Vision Language Models (VLMs) have proven effective for zero-shot classification. In this paper, we aim at a more challenging setting, Realistic Zero-Shot Classification, which assumes no annotation but instead a broad vocabulary. We propose the Self Structural Semantic Alignment (S3A) framework, which extracts structural semantic information from unlabeled data while simultaneously self-learning.
  arXiv  Detail & Related papers  (2023-08-24T17:56:46Z)
• Understanding Imbalanced Semantic Segmentation Through Neural Collapse [81.89121711426951]
  We show that semantic segmentation naturally brings contextual correlation and imbalanced distribution among classes. We introduce a regularizer on feature centers to encourage the network to learn features closer to the appealing structure. Our method ranks 1st and sets a new record on the ScanNet200 test leaderboard.
  arXiv  Detail & Related papers  (2023-01-03T13:51:51Z)
• Unsupervised Representation Learning for 3D Point Cloud Data [66.92077180228634]
  We propose a simple yet effective approach for unsupervised point cloud learning. In particular, we identify a very useful transformation which generates a good contrastive version of an original point cloud. We conduct experiments on three downstream tasks which are 3D object classification, shape part segmentation and scene segmentation.
  arXiv  Detail & Related papers  (2021-10-13T10:52:45Z)

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.