# Class-Imbalanced Semi-Supervised Learning for Large-Scale Point Cloud Semantic Segmentation via Decoupling Optimization Source: [Class-Imbalanced Semi-Supervised Learning for Large-Scale Point Cloud Semantic Segmentation via Decoupling Optimization]() Collection index: [llms.txt]() Data updated: 2025-10-26T02:12:32.787Z Authors: Mengtian Li, Shaohui Lin, Zihan Wang, Yunhang Shen, Baochang Zhang, Lizhuang Ma Published: 2024-01-13T04:16:40.000Z ## Abstract Semi-supervised learning (SSL), thanks to the significant reduction of data annotation costs, has been an active research topic for large-scale 3D scene understanding. However, the existing SSL-based methods suffer from severe training bias, mainly due to class imbalance and long-tail distributions of the point cloud data. As a result, they lead to a biased prediction for the tail class segmentation. In this paper, 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. In particular, we first employ two-round pseudo-label generation to select unlabeled points across head-to-tail classes. We further introduce multi-class imbalanced focus loss to adaptively pay more attention to feature learning across head-to-tail classes. We fix the backbone parameters after feature learning and retrain the classifier using ground-truth points to update its parameters. Extensive experiments demonstrate the effectiveness of our method outperforming previous state-of-the-art methods on both indoor and outdoor 3D point cloud datasets (i.e., S3DIS, ScanNet-V2, Semantic3D, and SemanticKITTI) using 1% and 1pt evaluation. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. ## Original sources and implementation [Original paper]() [PDF]()