Coordination of view coverage via privacy-aware smart cameras is key to a more socially responsible urban intelligence. Rather than maximizing view coverage at any cost or over relying on expensive cryptographic techniques, we address how cameras can coordinate to legitimately monitor public spaces while excluding privacy-sensitive regions by design. This article proposes a decentralized framework in which interactive smart cameras coordinate to autonomously select their orientation via collective learning, while eliminating privacy violations via soft and hard constraint satisfaction. The approach scales to hundreds up to thousands of cameras without any centralized control. Experimental evidence shows 18.42% higher coverage efficiency and 85.53% lower privacy violation than baselines and other state-of-the-art approaches. This significant advance further unravels practical guidelines for operators and policymakers: how the field of view, spatial placement, and budget of cameras operating by ethically-aligned artificial intelligence jointly influence coverage efficiency and privacy protection in large-scale and sensitive urban environments.
Figures & tables
Attributes
I-EPOS [ 7 , 8 ]
EPOS [ 9 ]
COHDA [ 10 ]
GGV [ 11 ]
Computational cost
agent: O(KL) ; system: O(KLlogU)
agent: O(KC) ; system: O(KClogU)
agent: O(KL) ; system: O(KL)
agent: O(KL) ; system: O(KLU)
Communication cost
agent: O(L) ; system: O(LlogU)
agent: O(K) ; system: O(KlogU)
agent: O(UL) ; system: O(UL)
agent: O(UL) ; system: O(UL)
Scalability
✓
✓
×
×
Privacy awareness
✓
×
×
×
K : number of plans (options), L : number of iterations, C : number of children, U : number of agents
Criteria covered ✓ or not ×
TABLE I: Comparison of approaches for discrete-choice combinatorial optimization problem.
Fig. 1: Overview of the smart camera coordination model. Camera A has three orientation plans ( 30∘ , 210∘ , and 300∘ ). When camera A rotates 30∘ , its field of view (FoV) covers 40% of cell a2 , 70% of cell a3 and 10% of cell a4 . It generates all possible 360∘ plans and then coordinates with other cameras to observe the required regions while excluding the private regions through collective learning. The objective is to match the overall coverage to the target at each cell.
Fig. 2: The coverage value per cell required by the targets with and without private regions.
Fig. 3: Coverage inefficiency of three approaches with different types of camera placement and number of plans per camera. The shadow represents the standard error of I-EPOS . The heatmaps show the coverage performance (match, overlap, or loss) on the 2D map through different methods.
Fig. 4: The coverage performance comparison of different approaches using 10×10 cameras (each with 90 plans) in the maps with three types of private regions. The metrics include coverage inefficiency, privacy violation rate, and total coverage ratio.
Fig. 5: Comparison of interpolated total cost and coverage performance among different types of camera placement. Each type of camera placement requires smart cameras with different horizontal angles and prices according to FIXr.
Video surveillance uses computer vision and multimodal AI to detect, track, explain, and retrieve events in camera streams while balancing reliability, efficiency, and privacy.
City-scale person re-identification across distributed cameras must handle severe appearance changes from viewpoint, occlusion, and domain shift while complying with data protection rules that prevent sharing raw imagery. We introduce CityGuard, a topology-aware transformer for privacy-preserving identity retrieval in decentralized surveillance. The framework integrates three components. A dispersion-adaptive metric learner adjusts instance-level margins according to feature spread, increasing intra-class compactness. Spatially conditioned attention injects coarse geometry, such as GPS or deployment floor plans, into graph-based self-attention to enable projectively consistent cross-view alignment using only coarse geometric priors without requiring survey-grade calibration. Differentially private embedding maps are coupled with compact approximate indexes to support secure and cost-efficient deployment. Together these designs produce descriptors robust to viewpoint variation, occlusion, and domain shifts, and they enable a tunable balance between privacy and utility under rigorous differential-privacy accounting. Experiments on Market-1501 and additional public benchmarks, complemented by database-scale retrieval studies, show consistent gains in retrieval precision and query throughput over strong baselines, confirming the practicality of the framework for privacy-critical urban identity matching.
With the increasing deployment of intelligent sensing technologies in highly sensitive environments such as restrooms and locker rooms, visual surveillance systems face a profound privacy-security paradox. Existing privacy-preserving approaches, including physical desensitization, encryption, and obfuscation, often compromise semantic understanding or fail to ensure mathematically provable irreversibility. Although Privacy Camera 1.0 eliminated visual data at the source to prevent leakage, it provided only textual judgments, leading to evidentiary blind spots in disputes. To address these limitations, this paper proposes a novel privacy-preserving perception framework based on the AI Flow paradigm and a collaborative edge-cloud architecture. By deploying a visual desensitizer at the edge, raw images are transformed in real time into abstract feature vectors through nonlinear mapping and stochastic noise injection under the Information Bottleneck principle, ensuring identity-sensitive information is stripped and original images are mathematically unreconstructable. The abstract representations are transmitted to the cloud for behavior recognition and semantic reconstruction via a "dynamic contour" visual language, achieving a critical balance between perception and privacy while enabling illustrative visual reference without exposing raw images.
Huan Song, Shuyu Tian, Ting Long +5
Institute of Artificial Intelligence (TeleAI), China Telecom
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.
David Voihanski, Mor Sinai, Ben Zion Bobrovsky
Juganu, Or Yehuda, Israel · Tel Aviv University, Israel