Jincheng Xu

Open-vocabulary object detection in unmanned aerial vehicle (UAV) imagery remains challenging under zero-shot cross-dataset transfer because tiny and cluttered targets are highly sensitive to early resolution reduction under domain shift. This study aims to improve transferable open-vocabulary UAV detection by revisiting stride-2 downsampling in YOLO-World v2 as a transfer-critical bottleneck. AeroPinWorld is introduced as a pinwheel-augmented YOLO-World v2 that selectively replaces four key stride-2 transitions with pinwheel-shaped convolution (PConv) so as to reduce aliasing, alleviate sampling-phase sensitivity, and preserve fine-grained local structures, while keeping the original detection head unchanged to ensure a fair and efficient comparison. The model is trained on COCO2017 for 24 epochs from official pretrained weights and directly evaluated, without target-domain fine-tuning, on VisDrone2019-DET and UAVDT using fixed offline prompt vocabularies. Compared with YOLO-World v2-S, AeroPinWorld improves zero-shot transfer performance on VisDrone from 0.112 to 0.135 mAP and from 0.054 to 0.063 AP_S, and it also yields consistent gains on UAVDT. Ablation studies show that both early backbone replacements and head bottom–up replacements contribute to the final gains, with their combination achieving the best accuracy–efficiency trade-off. These results indicate that selectively redesigning transfer-critical downsampling operators is an effective and lightweight way to improve zero-shot open-vocabulary UAV detection under aerial domain shift.

Open-vocabulary object detectors enable prompt-driven recognition, yet their zero-shot transfer to unmanned aerial vehicle (UAV) imagery remains fragile under domain shift, where tiny and cluttered targets depend on weak fine-grained cues. We propose AeroPinWorld, a pinwheel-augmented YOLO-World v2 that revisits stride-2 downsampling as a key transfer bottleneck: aggressive resolution reduction can induce aliasing-driven detail loss and sampling-phase sensitivity, which disproportionately harms small-object representations and degrades cross-dataset generalization in aerial scenes. To address this, AeroPinWorld introduces pinwheel-shaped convolution (PConv) as a phase-aware reduction operator. PConv probes complementary offsets via asymmetric padding and directional kernels before feature fusion, strengthening local structure aggregation at downsampling junctions. Importantly, we do not replace all downsampling operations; instead, we selectively substitute PConv at critical pyramid transitions, including the first two backbone reductions (P1/2 and P2/4) and the two bottom-up stride-2 reductions in the head, while keeping later backbone stages unchanged to preserve efficiency. We evaluate under a strict zero-shot cross-dataset protocol by training on COCO2017 for 24 epochs from official pretrained weights and directly testing on two UAV benchmarks, VisDrone2019-DET and UAVDT, without any target-domain fine-tuning, using an offline prompt vocabulary at inference. Experiments demonstrate consistent improvements over the baseline, achieving a +2.3 mAP gain and a +0.9 APS gain on VisDrone, and yielding consistent gains on UAVDT, while maintaining a competitive efficiency profile.