SCOPE: Scene-Contextualized Incremental Few-Shot 3D Segmentation

2026-03-06

AI Summary

This work addresses catastrophic forgetting and insufficient prototype discriminability in incremental few-shot 3D point cloud segmentation, which stem from sparse annotations. It is the first to exploit the key observation that novel classes often appear as unlabeled background within base-class scenes. To this end, the authors propose a plug-and-play background-guided prototype enhancement framework. After base-class training, high-confidence pseudo-instances are extracted from the background to construct a prototype pool; during novel-class learning, these background prototypes are fused with few-shot prototypes—without retraining the backbone or introducing additional parameters. Experiments show consistent improvements: on ScanNet and S3DIS, the method boosts novel-class IoU by 6.98% and 3.61%, respectively, and average IoU by 2.25% and 1.70%, achieving state-of-the-art performance while significantly mitigating catastrophic forgetting.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

Incremental Few-Shot (IFS) segmentation aims to learn new categories over time from only a few annotations. Although widely studied in 2D, it remains underexplored for 3D point clouds. Existing methods suffer from catastrophic forgetting or fail to learn discriminative prototypes under sparse supervision, and often overlook a key cue: novel categories frequently appear as unlabelled background in base-training scenes. We introduce SCOPE (Scene-COntextualised Prototype Enrichment), a plug-and-play background-guided prototype enrichment framework that integrates with any prototype-based 3D segmentation method. After base training, a class-agnostic segmentation model extracts high-confidence pseudo-instances from background regions to build a prototype pool. When novel classes arrive with few labelled samples, relevant background prototypes are retrieved and fused with few-shot prototypes to form enriched representations without retraining the backbone or adding parameters. Experiments on ScanNet and S3DIS show that SCOPE achieves SOTA performance, improving novel-class IoU by up to 6.98% and 3.61%, and mean IoU by 2.25% and 1.70%, respectively, while maintaining low forgetting. Code is available https://github.com/Surrey-UP-Lab/SCOPE.

Problem

• Incremental Few-Shot Segmentation
• 3D Point Clouds
• Catastrophic Forgetting
• Prototype Learning
• Sparse Supervision

Innovation

• incremental few-shot learning
• 3D point cloud segmentation
• prototype enrichment
• background pseudo-labeling
• scene context