Xuelin Qian

UniRef-UAV: A Multimodal Benchmark for Universal Referring in UAV Imagery

Haibin Tian, Huichao Xie, Xuelin Qian, Ruitao Lu, Junwei Han, Dingwen Zhang

Unmanned aerial vehicles (UAVs) increasingly rely on visual grounding capabilities to localize task-relevant targets from diverse instructions in complex aerial scenes. Existing referring expression comprehension (REC) benchmarks and methods, however, are largely built around tex…Read more