PulseAugur
EN
LIVE 12:31:10

New AI methods enhance pedestrian privacy in autonomous vehicle datasets

By PulseAugur Editorial · [4 sources] ·

Researchers have developed new methods for protecting pedestrian privacy in datasets used for autonomous vehicles. One approach, detailed in a five-stage pipeline, uses face-swapping models like Roop to conceal identities while preserving essential facial attributes for training AI models. Another method, LPID, introduces imperceptible perturbations to images that prevent unauthorized face-recognition models from learning from them, even when faces are extracted and resized. AI

IMPACT These privacy-preserving techniques could enable the creation of larger, more diverse datasets for autonomous vehicle training without compromising individual privacy.

RANK_REASON The cluster contains two research papers published on arXiv detailing novel methods for AI-driven privacy protection.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

New AI methods enhance pedestrian privacy in autonomous vehicle datasets

COVERAGE [4]

  1. arXiv cs.AI TIER_1 English(EN) · Roba H. Farouk, Catherine M. Elias ·

    Swapping Faces, Saving Features: A Dual-Purpose Pipeline for Pedestrian Privacy in ITS

    arXiv:2607.08402v1 Announce Type: cross Abstract: Large-scale and diverse datasets are needed to train AI models to take real-time decisions for autonomous vehicles (AVs), an intelligent transportation system (ITS) application. Pedestrian intention and trajectory prediction are c…

  2. arXiv cs.AI TIER_1 English(EN) · Catherine M. Elias ·

    Swapping Faces, Saving Features: A Dual-Purpose Pipeline for Pedestrian Privacy in ITS

    Large-scale and diverse datasets are needed to train AI models to take real-time decisions for autonomous vehicles (AVs), an intelligent transportation system (ITS) application. Pedestrian intention and trajectory prediction are critical models used in AVs, requiring datasets inv…

  3. arXiv cs.CV TIER_1 English(EN) · Byunghoon Oh, Sunghwan Park, Jaewoo Lee ·

    Unlearnable Faces: Privacy Protection Surviving Extraction Pipeline

    arXiv:2607.05996v1 Announce Type: new Abstract: Unlearnable examples keep publicly shared photos from being learned by unauthorized face-recognition models. An imperceptible perturbation, added before sharing, makes any model trained on the protected photos fail on clean faces. T…

  4. arXiv cs.CV TIER_1 English(EN) · Jaewoo Lee ·

    Unlearnable Faces: Privacy Protection Surviving Extraction Pipeline

    Unlearnable examples keep publicly shared photos from being learned by unauthorized face-recognition models. An imperceptible perturbation, added before sharing, makes any model trained on the protected photos fail on clean faces. The perturbation is crafted on the shared image, …

RELATED ENTITIES

RELATED TOPICS