Weakly Supervised Fine-Grained Aircraft Detection Based on Prior-Knowledge Prototype Learning
Published in MDPI Remote Sensing, 2026
This paper proposes a two-stage weakly supervised fine-grained aircraft detection framework that eliminates reliance on large-scale fine-grained annotations. It introduces a Prior-knowledge-based Soft-label Generator (PSG) to propagate fine-grained labels, a Text Prototype Alignment (TPA) strategy to enhance semantic discrimination, and a Feature Purification Module (FPM) to suppress background noise, achieving significant performance gains with only 1% labeled data.
Recommended citation:
P. Chen,Y. Qiu, R. Li and Z. Zhu (corresponding author), "Weakly Supervised Fine-Grained Aircraft Detection Based on Prior-Knowledge Prototype Learning," in Remote Sensing, vol. 18, p. 2390, 2026, doi: 10.3390/rs18142390.
Download Paper