VPD-100K: Towards Generalizable and Fine-grained Visual Privacy Protection

TL;DR

This paper introduces VPD-100K, a large-scale, fine-grained dataset for visual privacy detection, along with a novel frequency-enhanced module to improve detection accuracy in real-world scenarios.

Contribution

The paper provides a comprehensive privacy dataset with detailed annotations and proposes a new frequency-domain attention mechanism to enhance privacy detection models.

Findings

VPD-100K contains 100,000 images with 33 fine-grained privacy classes.

The frequency-enhanced module improves detection performance on image and video benchmarks.

Extensive experiments validate the dataset's usefulness and the effectiveness of the proposed method.

Abstract

Privacy protection has become a critical requirement in the era of ubiquitous visual data sharing, imposing higher demands on efficient and robust privacy detection algorithms. However, current robust detection models are severely hindered by the lack of comprehensive datasets. Existing privacy-oriented datasets often suffer from limited scale, coarse-grained annotations, and narrow domain coverage, failing to capture the intricate details of sensitive information in realworld environments. To bridge this gap, we present a large-scale, fine-grained Visual Privacy Dataset (VPD-100K), designed to facilitate generalized privacy detection. We establish a holistic taxonomy comprising four primary domains: Human Presence, On-Screen Personally Identifiable Information (PII), Physical Identifiers, and Location Indicators, containing 100,000 images annotated with 33 fine-grained classes and over 190,000 object instances. Statistical analysis reveals that our dataset features long-tailed distributions, small object scales, and high visual complexity. These characteristics make the dataset particularly valuable for demanding, unconstrained applications such as live streaming, where actors frequently face unintentional, realtime information leakage. Furthermore, we design an effective frequency-enhanced lightweight module consisting of frequency-domain attention fusion and adaptive spectral gating mechanism that breaks the limitations of spatial pixel intensity to better capture the subtle details of sensitive information. Extensive experiments conducted on both diverse image and streaming videos benchmarks consistently demonstrate the effectiveness of our VPD-100K dataset and the wellcurated frequency mechanism. The code and dataset are available at https://vpd-100k.github.io/.