AI RESEARCH
Cross-modal Fuzzy Alignment Network for Text-Aerial Person Retrieval and A Large-scale Benchmark
arXiv CS.CV
•
ArXi:2603.20721v1 Announce Type: new Text-aerial person retrieval aims to identify targets in UAV-captured images from eyewitness descriptions, ing intelligent transportation and public security applications. Compared to ground-view text--image person retrieval, UAV-captured images often suffer from degraded visual information due to drastic variations in viewing angles and flight altitudes, making semantic alignment with textual descriptions very challenging.