Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.13263 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912833513455616 |
|---|---|
| author | Liu, Chenyu Cecotti, Marco Vijayakumar, Harikrishnan Robinson, Patrick Barson, James Caleap, Mihai |
| author_facet | Liu, Chenyu Cecotti, Marco Vijayakumar, Harikrishnan Robinson, Patrick Barson, James Caleap, Mihai |
| contents | Developing cost-efficient and reliable perception systems remains a central challenge for automated vehicles. LiDAR and camera-based systems dominate, yet they present trade-offs in cost, robustness and performance under adverse conditions. This work introduces a novel framework for learning-based 3D semantic segmentation using Calyo Pulse, a modular, solid-state 3D ultrasound sensor system for use in harsh and cluttered environments. A 3D U-Net architecture is introduced and trained on the spatial ultrasound data for volumetric segmentation. Results demonstrate robust segmentation performance from Calyo Pulse sensors, with potential for further improvement through larger datasets, refined ground truth, and weighted loss functions. Importantly, this study highlights 3D ultrasound sensing as a promising complementary modality for reliable autonomy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_13263 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Deep Learning for Semantic Segmentation of 3D Ultrasound Data Liu, Chenyu Cecotti, Marco Vijayakumar, Harikrishnan Robinson, Patrick Barson, James Caleap, Mihai Computer Vision and Pattern Recognition Developing cost-efficient and reliable perception systems remains a central challenge for automated vehicles. LiDAR and camera-based systems dominate, yet they present trade-offs in cost, robustness and performance under adverse conditions. This work introduces a novel framework for learning-based 3D semantic segmentation using Calyo Pulse, a modular, solid-state 3D ultrasound sensor system for use in harsh and cluttered environments. A 3D U-Net architecture is introduced and trained on the spatial ultrasound data for volumetric segmentation. Results demonstrate robust segmentation performance from Calyo Pulse sensors, with potential for further improvement through larger datasets, refined ground truth, and weighted loss functions. Importantly, this study highlights 3D ultrasound sensing as a promising complementary modality for reliable autonomy. |
| title | Deep Learning for Semantic Segmentation of 3D Ultrasound Data |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.13263 |