Abstract
Automatic building footprint extraction from SAR imagery is one of the critical tasks in the remote sensing community. CNN has been recently explored in building extraction tasks and achieved improved performance. However, due to the scarcity of training data, it suffers from overfitting problem. This paper presents a novel knowledge distillation based framework consisting of teacher and student networks. Regarding EO image as privileged information, the teacher network learns to extract the rich EO and SAR image pair features. The student network then learns to estimate the building footprints from only SAR images based on the privileged knowledge from the teacher network. Experimental results on SpaceNet-6 benchmark demonstrate the effectiveness of our framework, which explicitly improves the performance of SAR segmentation network.
| Original language | English |
|---|---|
| Pages | 3014-3017 |
| Number of pages | 4 |
| DOIs | |
| Publication status | Published - 2021 |
| Event | 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021 - Brussels, Belgium Duration: 2021 Jul 12 → 2021 Jul 16 |
Conference
| Conference | 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021 |
|---|---|
| Country/Territory | Belgium |
| City | Brussels |
| Period | 21/7/12 → 21/7/16 |
Bibliographical note
Funding Information:∗Corresponding author. Email: khsohn@yonsei.ac.kr †This research was supported by the Agency for Defense Development under the grant UD2000008RD.
Funding Information:
Acknowledge. This research was supported by the Yonsei University Research Fund of 2021 (2021-22-0001).
Publisher Copyright:
© 2021 IEEE.
All Science Journal Classification (ASJC) codes
- Computer Science Applications
- Earth and Planetary Sciences(all)