# Sarus: Privacy-Preserving Multi-Vendor Perception Fusion via Homomorphic Encryption

arXiv: https://arxiv.org/abs/2607.19146 · HTML: https://mlsanity.com/paper/2607.19146/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 53 references that are in the map.
- First posted on arXiv 21 Jul 2026.

## Builds on (most cited first, 21 of 21)

- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [Advances and Open Problems in Federated Learning](https://mlsanity.com/paper/1912.04977/) (2019)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection](https://mlsanity.com/paper/1912.13192/) (2019)
- [V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer](https://mlsanity.com/paper/2203.10638/) (2022)
- [OPV2V: An Open Benchmark Dataset and Fusion Pipeline for Perception with Vehicle-to-Vehicle Communication](https://mlsanity.com/paper/2109.07644/) (2021)
- [V2VNet: Vehicle-to-Vehicle Communication for Joint Perception and Prediction](https://mlsanity.com/paper/2008.07519/) (2020)
- [Cooper: Cooperative Perception for Connected Autonomous Vehicles based on 3D Point Clouds](https://mlsanity.com/paper/1905.05265/) (2019)
- [Cooperative Perception for 3D Object Detection in Driving Scenarios using Infrastructure Sensors](https://mlsanity.com/paper/1912.12147/) (2019)
- [Federated Learning: Opportunities and Challenges](https://mlsanity.com/paper/2101.05428/) (2021)
- [CryptoDL: Deep Neural Networks over Encrypted Data](https://mlsanity.com/paper/1711.05189/) (2017)
- [Learning for Vehicle-to-Vehicle Cooperative Perception under Lossy Communication](https://mlsanity.com/paper/2212.08273/) (2022)
- [CryptoNN: Training Neural Networks over Encrypted Data](https://mlsanity.com/paper/1904.07303/) (2019)
- [MACP: Efficient Model Adaptation for Cooperative Perception](https://mlsanity.com/paper/2310.16870/) (2023)
- [Inferring Spatial Uncertainty in Object Detection](https://mlsanity.com/paper/2003.03644/) (2020)
- [Vehicular Cooperative Perception Through Action Branching and Federated Reinforcement Learning](https://mlsanity.com/paper/2012.03414/) (2020)
- [On the Federated Learning Framework for Cooperative Perception](https://mlsanity.com/paper/2404.17147/) (2024)
- [Hermes Seal: Zero-Knowledge Assurance for Autonomous Vehicle Communications](https://mlsanity.com/paper/2603.26343/) (2026)
- [RG-Attn: Radian Glue Attention for Multi-modality Multi-agent Cooperative Perception](https://mlsanity.com/paper/2501.16803/) (2025)
- [On the Assessment of Sensitivity of Autonomous Vehicle Perception](https://mlsanity.com/paper/2602.00314/) (2026)

## Abstract (arXiv)

Cooperative perception enables autonomous vehicles (AVs) to improve situational awareness by aggregating detection outputs from multiple agents and sensing platforms, often via a shared fusion service in multi-vendor deployments. However, sharing such outputs at inference time exposes proprietary model behavior and sensitive environmental information, creating significant privacy and security concerns. In this paper, we present Sarus, a privacy-preserving framework for multi-vendor perception fusion via homomorphic encryption (HE), enabling aggregation without revealing individual vendor outputs. Each vendor encodes detections as compact Gaussian moment vectors over a shared spatial lattice and transmits encrypted payloads to a fusion server, which aggregates them directly in the encrypted domain. The fused result is then decrypted and reconstructed into final detections through class-wise bin merging. We analyze the computational complexity, showing linear scaling for vendor payload construction and O(BV) server-side fusion with the number of occupied bins B and vendors V, while postprocessing scales as O(B + sumc∈ C Bc2), where C denotes the set of object classes and Bc is the number of occupied bins for class c. Experiments demonstrate linear scaling in practice with only a bounded constant-factor overhead from HE, with decryption dominating postprocessing cost. Experiments on the KITTI dataset using camera (YOLOv8) and LiDAR (PointPillars, PV-RCNN) detectors show that Sarus improves scene-level coverage by effectively aggregating complementary detections, particularly in distance-dependent regimes where individual modalities degrade. These results indicate that privacy-preserving multi-vendor perception fusion is feasible for real-time deployment when statistical compression and spatial sparsity are jointly exploited.