Validated Adaptation for Aerial Crowd Monitoring at Mass Gathering Scale: A Deployment Protocol, a Severity Law, and a Diagnostic for Label-Free Drone Crowd Counting, Toward the FIFA World Cup 2034 (Saudi Arabia)

AlAnoud AllGhayth; AlJawharh AlOtaibi; Jude AlSubaie

cs.AIAug 18, 2026

Validated Adaptation for Aerial Crowd Monitoring at Mass Gathering Scale: A Deployment Protocol, a Severity Law, and a Diagnostic for Label-Free Drone Crowd Counting, Toward the FIFA World Cup 2034 (Saudi Arabia)

Authors: AlAnoud AllGhayth, AlJawharh AlOtaibi, Jude AlSubaie

Organizations: Riyadh, Saudi Arabia

Abstract

Saudi Arabia will host the 2034 FIFA World Cup and already operates crowd management at Hajj scale. Drone-based counting must hold accuracy on footage unlike anything in its training corpus, without labels, and must warn of dangerous inflow before a crush forms. We deliver a validated answer built on 525 controlled runs, a full-resolution corpus study, five falsification ablations, and a five-condition safety-interlock evaluation. Label-free adaptation recovers 31-49% of shift-induced error across four corruptions and five severities, with the strongest method gaining 41.8 MAE over the frozen source (95% CI [34.1, 49.6], p=7.5x10^-10, d=2.52). We establish a severity law separating methods with a constant absolute margin from the one whose margin grows, and a stability budget identifying which configuration is safe to fly. On a full-resolution corpus carrying a genuine +48 MAE aerial gap (source retrained to 14.6 validation MAE, a 34% improvement), adaptation repairs the dense-scene undercounting that would otherwise under-report a forming crush, and the flux-based risk module fires on real congestion episodes in 2 of 6 full-length clips. We localise the recoverable error: in a regime built to favor a physics-informed conservation prior (300-frame clips at 200ms spacing, five times wider than standard), the adaptation signal is normalisation-driven, not flow-driven; the continuity residual is invariant to the proportional counting errors domain shift produces, confirmed by four on/off ablations correlated at r=0.999 and a 40% input corruption moving accuracy by only 0.05 MAE. A label-free shift gate shows shift magnitude and accuracy damage are rank-independent (Spearman rho=0.20; rho=-0.60 among genuine shifts), quantifying the 58% of headroom a magnitude gate forgoes. We establish unconditional adaptation with tail monitoring as policy, closing with a six-point protocol.

Figures & tables

	Track A	Track B
Purpose	controlled shift	real domain gap
Corpus	subset, $n{=}750$	full release, full-res
Clip length	short	300 frames
Frame spacing	40 ms	${\sim}200$ ms
Source val mae	26.1	14.6
Runs	525	ablations + risk

Table 1: The three experimental tracks. All adaptation is label-free and updates only BN parameters. Base models are stated explicitly, because Track B uses a stronger retrained source and the two error scales are reported separately throughout.

Figure 1: Adaptation across corruption severity. (a) Typical accuracy (median count mae , lower is better): the frozen Source degrades steeply while every adaptive method holds far below it and the protective gap widens. (b) Stability (worst replicate, min–max band): TENT+Ours carries the extreme tail, peaking near $113$ mae at motion-blur severity 1 (best typical accuracy, widest spread). Panel (b) is the basis for the stability budget in Section 10 .

Density band	Source	Adapt	Source	Adapt	$n$
	mae		Bias
Dense ( $>291$ )	194.7	98.5	$-194.7$	$-98.4$	605
Medium ( $129$ – $291$ )	170.1	96.2	$-170.1$	$-96.2$	591
Sparse ( $\leq 129$ )	70.3	9.4	$-70.3$	$-8.3$	598

Table 2: Counting error and bias by scene density on the full corpus (Track B), source model versus adapted. Bias is mean (predicted

-

true); negative is undercounting. Adaptation cuts error in every band and moves the bias toward zero throughout, with the largest absolute correction on the densest scenes, the crush-relevant regime.

Track	Condition	ON	OFF	$\Delta$
A	motion blur (sev. 2)	44.49	44.36	$+0.13$
A	low light	34.84	34.66	$+0.18$
B	clean ( $+48$ gap)	52.41	52.13	$+0.28$
B	low light	65.94	65.98	$-0.05$
Input-corruption ablation (Track A): clean flow $43.86$
vs. 40% corrupted flow $43.91$ ( $\Delta=+0.05$ mae ).

Table 3: Conservation on/off across four regimes.

\Delta

is mae (physics on)

-

mae (physics off). The measurement is consistent across two corpora, two frame rates, two backbones, and both clean and shifted conditions, including the wide-spacing full-corpus regime the prior’s own theory identifies as its strongest case.

Condition	$s$	fires	Source	Adapt	Gate
none	0.00110	no	36.9	34.6	36.9
JPEG	0.00135	no	87.5	44.3	87.5
motion blur	0.00207	no	93.4	42.2	93.4
Gaussian	0.00655	yes	102.3	67.1	67.1
low light	0.04279	yes	81.2	46.4	46.4
mean			80.3	46.9	66.3

Table 4: Shift-gated policy. The gate fires when the label-free shift score exceeds

\tau=2s_{\mathrm{clean}}=0.0022

. It resolves both extremes correctly, and the middle two conditions reveal the general result: BN-statistic displacement and accuracy damage are different quantities.

Figure 2: The shift-gated policy, decomposed. (a) The label-free shift score against the decision threshold $\tau=2s_{\mathrm{clean}}$ , annotated with the error that adaptation could recover in each condition. (b) What each policy delivered against what was available. The ordering of the bars in (a) and the ordering of the gains in (b) are close to independent (Spearman $\rho=0.20$ ; among the four genuine shifts, $\rho=-0.60$ ): statistical displacement and accuracy damage are different quantities, which is the general result of Section 8 .

Condition	Source	AdaBN	TENT	Ours (phys.)	TENT+Ours
Clean	$36.92\pm 0.00$	$33.77\pm 17.69$	$33.66\pm 17.41$	$\mathbf{32.93}\pm 15.50$	$57.12\pm 55.42$
Gaussian noise	$101.15\pm 0.00$	$73.66\pm 5.35$	$76.57\pm 5.28$	$78.83\pm 5.30$	$\mathbf{65.49}\pm 7.32$
Motion blur	$115.80\pm 0.00$	$48.71\pm 5.60$	$49.62\pm 6.47$	$51.35\pm 7.18$	$\mathbf{48.62}\pm 16.09$
Low light	$86.96\pm 0.00$	$46.68\pm 7.48$	$49.18\pm 7.96$	$51.46\pm 7.98$	$\mathbf{45.86}\pm 9.91$
JPEG	$80.44\pm 0.00$	$47.91\pm 5.46$	$49.34\pm 5.81$	$50.95\pm 6.16$	$\mathbf{47.58}\pm 7.50$
Mean (4 shifts)	$96.09$	$54.24$	$56.18$	$58.15$	$\mathbf{51.89}$

Table 5: Track A, reference severity. mae (mean

\pm

std over 5 replicates). Every adaptive method beats Source on every condition. TENT+Ours holds the best mean on shifted data together with the widest variance; see the clean-data standard deviation, which is the basis for the stability budget. Lower is better; best per row in bold.

	Mean mae				TENT recovered		Ours recovered		TENT+Ours recovered
Severity	Source	TENT	Ours	TENT+Ours	abs.	%	abs.	%	abs.	%
1	73.6	37.9	38.9	43.2	35.7	48.5	34.7	47.2	30.4	41.3
2	87.5	46.4	48.1	45.2	41.2	47.0	39.5	45.1	42.4	48.4
3	96.1	56.2	58.1	51.9	39.9	41.5	37.9	39.5	44.2	46.0
4	104.9	68.0	70.0	59.3	37.0	35.2	34.9	33.3	45.7	43.5
5	112.0	76.7	78.6	66.1	35.3	31.5	33.4	29.8	45.8	40.9

Table 6: The severity law (

4

corruptions

\times

5

severities

\times

5

replicates

=400

runs), pooled over corruptions. Left: mean mae per method. Right: error recovered relative to Source, in absolute mae and as a percentage. Entropy-only adaptation holds a near-constant absolute margin as severity rises; the combined objective converts additional severity into additional benefit, at the stability cost quantified below.

Crowd Counting

Crowd counting estimates how many people appear in images or video, supporting occupancy monitoring, safety planning, and crowd management.

See all 23 papers

HAJJv2-CrowdCount: Zero-Shot Benchmark for Dense Crowd Counting
Jul 8, 2026Reem AlYabis, Fares AlTuwaim, AlJawharh AlOtaibi +1Crowd Counting Object Counting
Riyadh, Saudi Arabia
Generative Adversarial Perturbations with Cross-paradigm Transferability on Localized Crowd Counting
Mar 25, 2026Alabi Mehzabin Anisha, Guangjing Wang, Sriram ChellappanAdversarial Attacks Transfer-Based Adversarial Attacks
University of South Florida, Tampa, Florida, USA
When Does Small Data Work? Accuracy and Efficiency Trade-offs Between Tabular Foundation Models and Conventional Methods for Crowd-State Classification at Hajj and Umrah
Jul 4, 2026AlJawharh S. AlOtaibi, Mohamed Eltahir, Jude AlSubaieTabular Foundation Models Few-Shot Learning
Riyadh, Saudi Arabia

Jul 8, 2026J/K move · Enter open · S save

HAJJv2-CrowdCount: Zero-Shot Benchmark for Dense Crowd Counting

Reem AlYabis, Fares AlTuwaim, AlJawharh AlOtaibi, Mohamed Eltahir

Riyadh, Saudi Arabia

Crowd Counting Object Counting

Open paper AskPDF arXiv

Automated crowd counting in Hajj video is difficult not because current models lack capacity, but because the footage violates the assumptions those models were built on: cameras observe the crowd from steep, near-vertical angles, individuals occlude one another extensively, and a single frame can contain well over a thousand people. Benchmarks that test crowd counting in such an environment are either private or not detailed per second. We revisit the HAJJv2 dataset and contribute HAJJv2-CrowdCount: per-second human-annotated crowd counts for its testing videos. Using these annotations, we benchmark three recent zero-shot counting paradigms: an open-vocabulary detector (YOLO-World), a point-based counter (APGCC), and a promptable segmentation-based counter (SAM3Count). SAM3Count attains the lowest overall mean absolute error (MAE 70.4, 95% CI 56.0-86.1), ahead of YOLO-World (92.0) and APGCC (152.9). This ordering reverses, however, in the regime most relevant to deployment: on the densest frames, the detection- and segmentation-based counters both degrade sharply (MAE exceeding 300), while the point-based counter degrades far more gracefully (MAE 114.9). This inversion is decision-relevant for Hajj crowd management, where reliable counts are needed most precisely in the densest and most occluded scenes. The annotations are released to support reproduction and extension of these results.

Abstract

Crowd Counting

Explore similar work

HAJJv2-CrowdCount: Zero-Shot Benchmark for Dense Crowd Counting

Generative Adversarial Perturbations with Cross-paradigm Transferability on Localized Crowd Counting

When Does Small Data Work? Accuracy and Efficiency Trade-offs Between Tabular Foundation Models and Conventional Methods for Crowd-State Classification at Hajj and Umrah