Privacy Preserving Visual SLAM

Shibuya, Mikiya; Sumikura, Shinya; Satoh, Ken

doi:10.1007/978-3-030-58542-6_7

Cited by 16 publications

(14 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Proposed by Speciale et al [64], the line-cloud representation obfuscate 2D / 3D point locations in the map building process [18,19,60] without compromising the accuracy in localization. However, since the descriptors are unchanged, Chelani et al [7] showed that line-clouds are vulnerable to inversion attacks if the underlying point-cloud is recovered.…”

Section: Related Workmentioning

confidence: 99%

“…These methods include obfuscating keypoint locations by † Corresponding author. lifting them to lines that pass through the original points [19,60,64,65], or to affine subspaces with augmented adversarial feature samples [16] to increase the difficulty of recovering the original images. However, recent work [7] has demonstrated that the closest points between lines can yield a good approximation to the original points locations, allowing descriptor inversion.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning

Ng¹,

Kim²,

Lee³

et al. 2021

Preprint

View full text Add to dashboard Cite

In the light of recent analyses on privacy-concerning scene revelation from visual descriptors, we develop descriptors that conceal the input image content. In particular, we propose an adversarial learning framework for training visual descriptors that prevent image reconstruction, while maintaining the matching accuracy. We let a feature encoding network and image reconstruction network compete with each other, such that the feature encoder tries to impede the image reconstruction with its generated descriptors, while the reconstructor tries to recover the input image from the descriptors. The experimental results demonstrate that the visual descriptors obtained with our method significantly deteriorate the image reconstruction quality with minimal impact on correspondence matching and camera localization performance.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning

Ng¹,

Kim²,

Lee³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…They argued that line clouds preserve privacy as they prevent the approaches from [57,73] from being applicable, which would ensure that user-recorded scenes can be safely stored in the cloud. Follow-up work to this seminal paper showed how to adapt SLAM systems to integrate this idea into a SLAM system [70] and how to enable privacy-preserving image queries for localization [75] and privacy-preserving SfM [24]. [24,75] operate on 2D rather than 3D representations and replace each 2D image feature by a 2D line.…”

Section: Related Workmentioning

confidence: 99%

“…They showed that the resulting representation is unintelligible to humans, prevents a direct application of [57], and still enables accurate camera pose estimation. This idea of lifting points to lines was later adapted for privacy-perserving SLAM [70]. However, this paper shows that it is possible to (approximately) recover the original 3D point positions from a line cloud (cf .…”

Section: Introductionmentioning

confidence: 99%

How Privacy-Preserving are Line Clouds? Recovering Scene Details from 3D Lines

Chelani¹,

Kahl²,

Sattler³

2021

Preprint

View full text Add to dashboard Cite

Visual localization is the problem of estimating the camera pose of a given image with respect to a known scene. Visual localization algorithms are a fundamental building block in advanced computer vision applications, including Mixed and Virtual Reality systems. Many algorithms used in practice represent the scene through a Structure-from-Motion (SfM) point cloud and use 2D-3D matches between a query image and the 3D points for camera pose estimation. As recently shown, image details can be accurately recovered from SfM point clouds by translating renderings of the sparse point clouds to images. To address the resulting potential privacy risks for user-generated content, it was recently proposed to lift point clouds to line clouds by replacing 3D points by randomly oriented 3D lines passing through these points. The resulting representation is unintelligible to humans and effectively prevents point cloudto-image translation. This paper shows that a significant amount of information about the 3D scene geometry is preserved in these line clouds, allowing us to (approximately) recover the 3D point positions and thus to (approximately) recover image content. Our approach is based on the observation that the closest points between lines can yield a good approximation to the original 3D points. Code is available at https://github.com/kunalchelani/Line2Point.

show abstract

“…while protecting against input reconstruction attacks [19]. For these setups, the dominant idea is to transform the point cloud into 3D line cloud [20] which obfuscates the semantic structure of the scene while preserving utility for camera localization [21], SLAM [22], SfM [23] etc. In contrast, we focus on providing utility for perception tasks (classification, detection, segmentation etc.)…”

Section: Introductionmentioning

confidence: 99%

Learning to Censor by Noisy Sampling

Chopra¹,

Java²,

Singh³

et al. 2022

Preprint

View full text Add to dashboard Cite

Point clouds are an increasingly ubiquitous input modality and the raw signal can be efficiently processed with recent progress in deep learning. This signal may, often inadvertently, capture sensitive information that can leak semantic and geometric properties of the scene which the data owner does not want to share. The goal of this work is to protect sensitive information when learning from point clouds; by censoring the sensitive information before the point cloud is released for downstream tasks. Specifically, we focus on preserving utility for perception tasks while mitigating attribute leakage attacks. The key motivating insight is to leverage the localized saliency of perception tasks on point clouds to provide good privacy-utility trade-offs. We realise this through a mechanism called Censoring by Noisy Sampling (CBNS ), which is composed of two modules: i) Invariant Sampler: a differentiable point-cloud sampler which learns to remove points invariant to utility and ii) Noisy Distorter: which learns to distort sampled points to decouple the sensitive information from utility, and mitigate privacy leakage. We validate the effectiveness of CBNS through extensive comparisons with state-ofthe-art baselines and sensitivity analyses of key design choices. Results show that CBNS achieves superior privacy-utility trade-offs on multiple datasets.

show abstract

Privacy Preserving Visual SLAM

Cited by 16 publications

References 41 publications

NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning

NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning

How Privacy-Preserving are Line Clouds? Recovering Scene Details from 3D Lines

Learning to Censor by Noisy Sampling

Contact Info

Product

Resources

About