Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.18335 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918305427619840 |
|---|---|
| author | Fan, Zexia Chen, Yu Zhang, Qiquan Chen, Kainan Qian, Xinyuan |
| author_facet | Fan, Zexia Chen, Yu Zhang, Qiquan Chen, Kainan Qian, Xinyuan |
| contents | Sound source localization (SSL) demonstrates remarkable results in controlled settings but struggles in real-world deployment due to dual imbalance challenges: intra-task imbalance arising from long-tailed direction-of-arrival (DoA) distributions, and inter-task imbalance induced by cross-task skews and overlaps. These often lead to catastrophic forgetting, significantly degrading the localization accuracy. To mitigate these issues, we propose a unified framework with two key innovations. Specifically, we design a GCC-PHAT-based data augmentation (GDA) method that leverages peak characteristics to alleviate intra-task distribution skews. We also propose an Analytic dynamic imbalance rectifier (ADIR) with task-adaption regularization, which enables analytic updates that adapt to inter-task dynamics. On the SSLR benchmark, our proposal achieves state-of-the-art (SoTA) results of 89.0% accuracy, 5.3° mean absolute error, and 1.6 backward transfer, demonstrating robustness to evolving imbalances without exemplar storage. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_18335 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Analytic Incremental Learning For Sound Source Localization With Imbalance Rectification Fan, Zexia Chen, Yu Zhang, Qiquan Chen, Kainan Qian, Xinyuan Sound Artificial Intelligence Sound source localization (SSL) demonstrates remarkable results in controlled settings but struggles in real-world deployment due to dual imbalance challenges: intra-task imbalance arising from long-tailed direction-of-arrival (DoA) distributions, and inter-task imbalance induced by cross-task skews and overlaps. These often lead to catastrophic forgetting, significantly degrading the localization accuracy. To mitigate these issues, we propose a unified framework with two key innovations. Specifically, we design a GCC-PHAT-based data augmentation (GDA) method that leverages peak characteristics to alleviate intra-task distribution skews. We also propose an Analytic dynamic imbalance rectifier (ADIR) with task-adaption regularization, which enables analytic updates that adapt to inter-task dynamics. On the SSLR benchmark, our proposal achieves state-of-the-art (SoTA) results of 89.0% accuracy, 5.3° mean absolute error, and 1.6 backward transfer, demonstrating robustness to evolving imbalances without exemplar storage. |
| title | Analytic Incremental Learning For Sound Source Localization With Imbalance Rectification |
| topic | Sound Artificial Intelligence |
| url | https://arxiv.org/abs/2601.18335 |