SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud Segmentation
Due to the limited quantity and high cost of high-quality three-dimensional annotations, generalized zero-shot point cloud segmentation aims to transfer the knowledge of seen to unseen classes by leveraging semantic correlations to achieve generalization purposes. Existing generative point cloud sem...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/15/8359 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849407815955775488 |
|---|---|
| author | Yuyun Wei Meng Qi |
| author_facet | Yuyun Wei Meng Qi |
| author_sort | Yuyun Wei |
| collection | DOAJ |
| description | Due to the limited quantity and high cost of high-quality three-dimensional annotations, generalized zero-shot point cloud segmentation aims to transfer the knowledge of seen to unseen classes by leveraging semantic correlations to achieve generalization purposes. Existing generative point cloud semantic segmentation approaches rely on generators trained on seen classes to synthesize visual features for unseen classes in order to help the segmentation model gain the ability of generalization, but this often leads to a bias toward seen classes. To address this issue, we propose a semantic-guided adaptive bias calibration approach with a dual-branch network architecture. This network consists of a novel visual–semantic fusion branch alongside the primary segmentation branch to suppress the bias toward seen classes. Specifically, the visual–semantic branch exploits the visual–semantic relevance of the synthetic features of unseen classes to provide auxiliary predictions. Furthermore, we introduce an adaptive bias calibration module that dynamically integrates the predictions from both the main and auxiliary branches to achieve unbiased segmentation results. Extensive experiments conducted on standard benchmarks demonstrate that our approach significantly outperforms state-of-the-art methods on both seen and unseen classes, thereby validating the effectiveness of our approach. |
| format | Article |
| id | doaj-art-87c9ee646a0a413293123b457ea2ed1f |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-87c9ee646a0a413293123b457ea2ed1f2025-08-20T03:35:57ZengMDPI AGApplied Sciences2076-34172025-07-011515835910.3390/app15158359SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud SegmentationYuyun Wei0Meng Qi1School of Information Science and Engineering, Shandong Normal University, Jinan 250358, ChinaSchool of Information Science and Engineering, Shandong Normal University, Jinan 250358, ChinaDue to the limited quantity and high cost of high-quality three-dimensional annotations, generalized zero-shot point cloud segmentation aims to transfer the knowledge of seen to unseen classes by leveraging semantic correlations to achieve generalization purposes. Existing generative point cloud semantic segmentation approaches rely on generators trained on seen classes to synthesize visual features for unseen classes in order to help the segmentation model gain the ability of generalization, but this often leads to a bias toward seen classes. To address this issue, we propose a semantic-guided adaptive bias calibration approach with a dual-branch network architecture. This network consists of a novel visual–semantic fusion branch alongside the primary segmentation branch to suppress the bias toward seen classes. Specifically, the visual–semantic branch exploits the visual–semantic relevance of the synthetic features of unseen classes to provide auxiliary predictions. Furthermore, we introduce an adaptive bias calibration module that dynamically integrates the predictions from both the main and auxiliary branches to achieve unbiased segmentation results. Extensive experiments conducted on standard benchmarks demonstrate that our approach significantly outperforms state-of-the-art methods on both seen and unseen classes, thereby validating the effectiveness of our approach.https://www.mdpi.com/2076-3417/15/15/8359adaptive bias calibrationvisual–semantic contrastive learninggeneralized zero-shot semantic segmentation |
| spellingShingle | Yuyun Wei Meng Qi SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud Segmentation Applied Sciences adaptive bias calibration visual–semantic contrastive learning generalized zero-shot semantic segmentation |
| title | SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud Segmentation |
| title_full | SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud Segmentation |
| title_fullStr | SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud Segmentation |
| title_full_unstemmed | SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud Segmentation |
| title_short | SemABC: Semantic-Guided Adaptive Bias Calibration for Generative Zero-Shot Point Cloud Segmentation |
| title_sort | semabc semantic guided adaptive bias calibration for generative zero shot point cloud segmentation |
| topic | adaptive bias calibration visual–semantic contrastive learning generalized zero-shot semantic segmentation |
| url | https://www.mdpi.com/2076-3417/15/15/8359 |
| work_keys_str_mv | AT yuyunwei semabcsemanticguidedadaptivebiascalibrationforgenerativezeroshotpointcloudsegmentation AT mengqi semabcsemanticguidedadaptivebiascalibrationforgenerativezeroshotpointcloudsegmentation |