Decentralized Multi-View, Multi-Zone Tracking for Scalable Camera Networks

David Voihanski; Mor Sinai; Ben Zion Bobrovsky

cs.CVOct 20, 2026

Decentralized Multi-View, Multi-Zone Tracking for Scalable Camera Networks

Authors: David Voihanski, Mor Sinai, Ben Zion Bobrovsky

Organizations: Juganu, Or Yehuda, Israel · Tel Aviv University, Israel

Abstract

We present a decentralized multi-view tracking system for scalable camera networks. Single-view detections are projected to bird's-eye-view (BEV) coordinates and fused by independent multi-view trackers, each managing a dynamic spatial zone. To support scalability across spatially partitioned camera networks, we introduce a decentralized layer for trajectory unification, utilizing a locking protocol and periodic track-merging operations termed global collate. This process is supported by a delayed-output buffering mechanism that retroactively applies history corrections. The configurable delay window absorbs asynchronous coordination latency and helps stabilize cross-zone identity consistency. We detail the design of dynamic zone polygons, consensus-based coordination, and the validation framework, which together maintain global ID consistency with minimal accuracy loss compared to a single-instance baseline in the evaluated configurations.

Figures & tables

Fig. 1: High-level architecture. Single-view trackers (outer arcs) run per camera and report smoothed leg points in BEV coordinates. One or more multi-view trackers operate on BEV detections within their zone polygons. A decentralized layer maintains overlapping BEV zones and coordinates global collate operations to maintain consistent track IDs across zones.

Fig. 2: Decentralized multi-view tracker main loop. BEV detections are processed by the local BEV tracker. Tracks whose hit count lies in the global-collate window are collected, and the instance attempts to acquire the shared global lock for all overlapping zones. If the lock is obtained, it runs global collate and then collate validation; all accepted and reverted merges are written back into the match-history database before the iteration completes.

Fig. 3: Dynamic zone polygon expansion. The multi-view zone starts as a small polygon around the first few detections (left). As new detections arrive outside the current polygon, the polygon expands to include them (middle). After sufficient coverage, the polygon stabilizes and represents the final spatial footprint of the zone in bird’s-eye coordinates (right).

Fig. 4: Zone polygon management in the decentralized tracker. For each incoming BEV detection $(x,y)$ , the instance checks whether the point lies inside its current zone polygon. If not, it expands its polygon and updates the decentralized zone_id $\rightarrow$ polygon map. All instances then check for polygon intersections; when overlaps are detected, each instance updates its local overlap map so that neighboring zones are known for later global collate.

Panel of Figure 5 — Fig. 5: Impact of global collate on cross-zone tracking. (a) Without global collate, a person moving across overlapping zones generates fragmented local tracks with distinct IDs. (b) With global collate, local tracks are matched and merged, yielding a single consistent global ID for the entire trajectory.

Fig. 6: Global collate decision flow. Candidate matches must first pass a minimum hit-ratio check. If the incoming ID is unique locally, the merge is accepted immediately. If the ID already exists (conflict), the system forces the swap—removing the old ID—only if the match cost satisfies a strict force threshold.

Fig. 7: Collate validation logic. For a set of input tracks, the instance sends a global collate request and examines the returned matches. For each matched pair it accumulates hit counts and cost sums, builds an old_tid $\rightarrow$ new_tid map, and then compares the validated cost of the old track vs. the candidate new track. Depending on this comparison, the system either reverts the global collate match or swaps the local track ID and updates the matches database.

Dataset	Configuration	HOTA ( $\uparrow$ )	IDF1 ( $\uparrow$ )	IDSW ( $\downarrow$ )
Dataset A	1 Inst. (All)	0.911	0.962	4
	2 Inst. (Split)	0.897	0.953	4
	3 Inst. (Split)	0.885	0.942	8
Dataset B	1 Inst. (All)	0.810	0.906	24
Dataset B	2 Inst. (Split)	0.809	0.908	26

TABLE I: Accuracy Comparison at 5.0 s Delay. Note that the 3-instance partition substantially reduces local multi-view redundancy (to only 3–4 cameras per instance), resulting primarily in localized intra-zone tracking dropouts rather than decentralized cross-zone matching failures (see also Table VI ).

Dataset	Configuration	HOTA ( $\uparrow$ )	IDF1 ( $\uparrow$ )	IDSW ( $\downarrow$ )
Dataset A (2 Inst.)	No Global Collate	0.635	0.706	48
	w/o Response Delay	0.869	0.937	13
	Full System	0.897	0.953	4
Dataset B (2 Inst.)	No Global Collate	0.735	0.855	64
	w/o Response Delay	0.779	0.891	43
	Full System	0.809	0.908	26

TABLE II: Ablation Study on Decentralized Components (2 Instances)

Dataset	Delay (s)	HOTA ( $\uparrow$ )	IDF1 ( $\uparrow$ )	IDSW ( $\downarrow$ )
Dataset A	0.0	0.869	0.937	13
	1.0	0.887	0.947	10
	2.0	0.896	0.951	9
	3.0	0.897	0.953	4
	5.0	0.897	0.953	4
Dataset B	0.0	0.779	0.891	43

TABLE III: Delay Sensitivity for Two-Instance Decentralized Runs

Dataset	Proposals	Accepted	Skipped	Forced	Collision Rej.	Validation No-op	Reversions
Dataset A (2 Inst.)	81	36 (44%)	24 (30%)	0	1 (1%)	20 (25%)	0
Dataset A (3 Inst.)	163	74 (45%)	68 (42%)	6 (4%)	12 (7%)	3 (2%)	5
Dataset B (2 Inst.)	88	34 (39%)	44 (50%)	5 (6%)	5 (6%)	0	2

TABLE IV: Global Collate Proposal Outcomes at 5.0 s Delay

Dataset & Config	GT Tracks	Crossing GT	Total Cross.	Mean Overlap Fr.
Dataset A (2 Inst.)	62	28	137	$\sim$ 844
Dataset A (3 Inst.)	62	34	165	$\sim$ 1155
Dataset B (2 Inst.)	45	25	59	$\sim$ 547

TABLE V: Ground-Truth Boundary Traffic and Overlap Profile

Dataset & Config	System Total	Intra-Zone	Cross-Zone IDSW
Dataset & Config	IDSW	IDSW	Collate-Corr.	No-Match
Dataset A (2 Inst.)	4	2	0	2
Dataset A (3 Inst.)	8	6	0	2
Dataset B (2 Inst.)	26	26	0	0

TABLE VI: Hierarchical Identity Switch Attribution Breakdown

Fig. 8: Total processing latency vs. number of zones ( $N=40$ cameras). Data is generated with 100% overlap (worst-case), yet latency improves as zones are added.

Fig. 9: Latency breakdown (100 tracks, 40 cameras). As zones increase, core tracking time (blue) decreases faster than decentralized coordination (orange) increases.

Multi-Object Tracking

Multi-object tracking (MOT) links detections across video frames to maintain object identities through motion, occlusion, and changing viewpoints.

See all 193 papers

Fully Distributed Multi-View 3D Tracking in Real-Time
Jun 11, 2026Byron Hernandez, Fangyu Li, Aotian Wu +33D MOT Multi-Object Tracking
University of Florida, Gainesville, FL, USA · NVIDIA Corporation, Santa Clara, CA, USA
On-the-Fly Homographies Calibration for Multi-Camera Tracking
Sep 16, 2026David Voihanski, Mor Sinai, Ben Zion BobrovskyMulti-View Geometry Camera Calibration
Juganu Or Yehuda, Israel · Tel Aviv University Israel
GRACE: Geometry- and Ray-Aware Camera-Efficient Multi-View Pedestrian Tracking
Sep 15, 2026Taigo Sakai, Hiroki Kouno, Naoki Kato +1Bird's-Eye-View Representation 3D MOT
Meijo university

Jun 11, 2026J/K move · Enter open · S save

Fully Distributed Multi-View 3D Tracking in Real-Time

Byron Hernandez, Fangyu Li, Aotian Wu, Paul J. Shin+2

University of Florida, Gainesville, FL, USA · NVIDIA Corporation, Santa Clara, CA, USA

3D MOT Multi-Object Tracking

Open paper AskPDF arXiv

Multi-camera tracking with overlapping fields of view typically relies on centralized fusion, which creates computational bottlenecks that prevent deployment at scale. We present MV3DT, a fully distributed framework for real-time multi-view 3D tracking that achieves accurate identity propagation and occlusion recovery through peer-to-peer coordination, eliminating the need for central aggregation. Each camera node executes a lightweight modular pipeline comprising monocular 3D perception, distributed multi-view association, and collaborative fusion via lightweight messaging. MV3DT achieves 96.5% IDF1, 93.1% MOTA, and 94.6% MOTP on WILDTRACK, competitive with state-of-the-art centralized methods, and unprecedented 41.7% IDF1 and 50.9% MOTA on SCOUT while demonstrating superior scalability: sustaining 30 FPS on 100 cameras with <10ms inter-camera latency and only 2.2% communication overhead. MV3DT operates in a zero-shot regime given camera calibrations, requiring no scene-specific learning and making it directly deployable in new environments. These results establish MV3DT as a practical solution for real-time multi-view tracking in large-scale overlapping camera networks.

Abstract

Multi-Object Tracking

Explore similar work

Fully Distributed Multi-View 3D Tracking in Real-Time

On-the-Fly Homographies Calibration for Multi-Camera Tracking

GRACE: Geometry- and Ray-Aware Camera-Efficient Multi-View Pedestrian Tracking