Organizations: School of Automation, Southeast University, Nanjing, China · Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China · School of Computer Science and Engineering, University of New South Wales, Sydney, Australia · School of Sports Training, Tianjin University of Sport, Tianjin, China · Tianjin University, Tianjin, China · National University of Defense Technology, Changsha, China
Air-ground cross-view referring person detection is a necessary component in the language-to-perception-to-control chain of collective embodied intelligence, grounding a language command into the same physical target before ground and aerial agents can coordinate downstream actions. Existing referring expression comprehension and open-vocabulary grounding methods do not jointly account for cross-view identity consistency, making them insufficient for Air-Ground Cross-View Referring Person Detection (AGCV-RPD), which involves similar pedestrian distractors, weak aerial appearance cues, and cross-view identity consistency. To study this problem, we introduce Air-Ground Paired Identity-Aware Referring (A-PAIR), the first comprehensive AGCV-RPD benchmark, containing 22,137 cross-view referring samples. To construct A-PAIR efficiently, we propose Factorized Annotation and Referential Alignment (FARA), a semi-automatic annotation framework that generates factorized referring descriptions and identity-consistency supervision at reduced cost. We propose Identity-Consistent Referring Grounding (ICRG), a framework that combines factorized referential grounding, candidate-completeness supervision, and cross-view consistency calibration for joint air-ground pair selection. ICRG improves ground, aerial, and pair-level detection over strong baselines, increasing pair F1 from 16.65% to 22.28%. These results show that AGCV-RPD requires paired detection and identity-consistent reasoning.
Figures & tables
Figure 1: AGCV-RPD task. The model detects the same referred person in paired ground and aerial views.
Figure 2: FARA annotation pipeline. The pipeline constructs A-PAIR from paired G2APS observations with factorized descriptions and identity labels.
Dataset
Lang.
Aerial
Ground
Air-Ground Pair
Same-ID Boxes
ID Aux.
Pair Eval.
RefCOCO ( Yu et al. 2016 )
✓
-
✓
-
-
-
-
RefCOCO+ ( Yu et al. 2016 )
✓
-
✓
-
-
-
-
RefCOCOg ( Mao et al. 2016 )
✓
-
✓
-
-
-
-
RefDrone ( Sun et al. 2026 )
✓
✓
-
-
-
-
-
VisDrone ( Du et al. 2019 )
-
✓
-
-
-
-
-
UAVDT ( Du et al. 2018 )
-
✓
-
-
-
-
-
Table 1: Benchmark comparison. A-PAIR covers language grounding, paired air-ground views, same-identity boxes, and pair-level evaluation.
Split
Samples
Pairs
Gnd. img.
Aer. img.
Train
15,497
6,672
6,672
3,530
Val
2,213
279
279
69
Test
4,427
637
637
292
Total
22,137
7,588
7,588
3,891
Table 2: A-PAIR split statistics.
Figure 3: ICRG framework. It combines factorized grounding, candidate completion, and cross-view consistency.
Method
Ground
Aerial
Pair
F1 i
Acc i
F1 img
Acc img
F1 i
Acc i
F1 img
Acc img
Acc
F1
TransVG ( Deng et al. 2021 )
2.53
1.28
4.94
2.53
0.05
0.02
0.09
0.05
0.00
0.00
PropVG ( Dai et al. 2025 )
1.49
0.75
2.94
1.49
0.00
0.00
0.00
0.00
0.00
0.00
Vision-R1 ( Zhan et al. 2026 )
1.92
0.97
3.77
1.92
0.00
0.00
0.00
0.00
0.00
0.00
SOREC-PIZA ( Goto et al. 2025 )
13.96
7.50
24.50
13.96
7.43
3.86
13.84
7.43
1.78
3.51
GDINO-B ( Liu et al. 2024 )
26.74
15.44
42.20
26.74
12.02
6.39
21.46
12.02
5.74
10.85
Table 3: Main results on A-PAIR. All metrics are percentages; bold and italics mark the best and second-best results.
Fact.
Cand.
Cons.
Ground
Aerial
Pair
F1 i
F1 img
F1 i
F1 img
Acc
F1
-
-
-
7.32
13.65
19.91
33.21
1.83
3.60
✓
-
-
34.88
51.72
20.83
34.47
11.66
20.88
✓
✓
-
36.12
53.07
21.82
35.82
12.24
21.82
✓
✓
✓
35.17
52.04
21.89
35.92
12.54
22.28
Table 4: Ablation results on A-PAIR. Columns indicate enabled supervision signals under the same protocol.
Figure 4: Pair-level IoU-quality decomposition. Each cell reports the percentage of A-PAIR test samples in a ground-aerial IoU-quality bin.
Figure 5: Ground versus pair performance. Each point shows one method by Ground F1 i and Pair F1 on A-PAIR.
Sun Yat-sen University, China · Pazhou Lab (HuangPu), Guangdong, China · Guangdong Province Key Laboratory of Information Security Technology, Guangzhou, China +1
School of Artificial Intelligence, Anhui University · State Key Laboratory of Opto-Electronic Information Acquisition and Protection Technology, Anhui University, Hefei, China · Anhui Provincial Key Laboratory of Security Artificial Intelligence, Anhui University, Hefei, China +2