← Back to publications
2024 · Journal article · Published

SIRST-5K: Exploring Massive Negatives Synthesis with Self-supervised Learning for Robust Infrared Small Target Detection

Yahao Lu, Yupei Lin, Han Wu, Xiaoyu Xian, Yukai Shi, Liang Lin
IEEE Transactions on Geoscience and Remote Sensing, 2024
Figure 1 — SIRST-5K: Exploring Massive Negatives Synthesis with Self-supervised Learning for Robust Infrared Small Target Detection
Fig. 1. Self-supervised strategy based on negative augmentation. To explore target correspondence in SIRST, we propose a negative augmentation approach to generate massive negatives for self-supervised representation learning.
Overview

Proposes a self-supervised negative-sample augmentation method and the SIRST-5K dataset to improve the diversity and robustness of infrared small-target detection.

Paper / Publisher ↗ PDF ↗ Code / Project ↗