# Beyond Anonymization: Object Scrubbing for Privacy-Preserving 2D and 3D Vision Tasks

arXiv: https://arxiv.org/abs/2504.16557 · HTML: https://mlsanity.com/paper/2504.16557/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 44 of its 62 references that are in the map.
- First posted on arXiv 23 Apr 2025.

## Builds on (most cited first, 30 of 44)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Masked-attention Mask Transformer for Universal Image Segmentation](https://mlsanity.com/paper/2112.01527/) (2021)
- [Membership Inference Attacks against Machine Learning Models](https://mlsanity.com/paper/1610.05820/) (2016)
- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [DETRs Beat YOLOs on Real-time Object Detection](https://mlsanity.com/paper/2304.08069/) (2023)
- [YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information](https://mlsanity.com/paper/2402.13616/) (2024)
- [Opacus: User-Friendly Differential Privacy Library in PyTorch](https://mlsanity.com/paper/2109.12298/) (2021)
- [Differentially Private Learning Needs Better Features (or Much More Data)](https://mlsanity.com/paper/2011.11660/) (2020)
- [NeuRAD: Neural Rendering for Autonomous Driving](https://mlsanity.com/paper/2311.15260/) (2023)
- [DP-CGAN: Differentially Private Synthetic Data and Label Generation](https://mlsanity.com/paper/2001.09700/) (2020)
- [GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators](https://mlsanity.com/paper/2006.08265/) (2020)
- [Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion](https://mlsanity.com/paper/2310.03502/) (2023)
- [DeepPrivacy: A Generative Adversarial Network for Face Anonymization](https://mlsanity.com/paper/1909.04538/) (2019)
- [Understanding Hallucinations in Diffusion Models through Mode Interpolation](https://mlsanity.com/paper/2406.09358/) (2024)
- [RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer](https://mlsanity.com/paper/2407.17140/) (2024)
- [Aggregated Contextual Transformations for High-Resolution Image Inpainting](https://mlsanity.com/paper/2104.01431/) (2021)
- [Torch.manual_seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision](https://mlsanity.com/paper/2109.08203/) (2021)
- [Natural and Effective Obfuscation by Head Inpainting](https://mlsanity.com/paper/1711.09001/) (2017)
- [DeepPrivacy2: Towards Realistic Full-Body Anonymization](https://mlsanity.com/paper/2211.09454/) (2022)
- [Defeating Image Obfuscation with Deep Learning](https://mlsanity.com/paper/1609.00408/) (2016)
- [A Hybrid Model for Identity Obfuscation by Face Replacement](https://mlsanity.com/paper/1804.04779/) (2018)
- [Attribute-preserving Face Dataset Anonymization via Latent Code Optimization](https://mlsanity.com/paper/2303.11296/) (2023)
- [Private Set Generation with Discriminative Information](https://mlsanity.com/paper/2211.04446/) (2022)
- [Qualitative Failures of Image Generation Models and Their Application in Detecting Deepfakes](https://mlsanity.com/paper/2304.06470/) (2023)
- [Combating Mode Collapse in GAN training: An Empirical Analysis using Hessian Eigenvalues](https://mlsanity.com/paper/2012.09673/) (2020)

## Abstract (arXiv)

We introduce ROAR (Robust Object Removal and Re-annotation), a scalable framework for privacy-preserving dataset obfuscation that eliminates sensitive objects instead of modifying them. Our method integrates instance segmentation with generative inpainting to remove identifiable entities while preserving scene integrity. Extensive evaluations on 2D COCO-based object detection show that ROAR achieves 87.5% of the baseline detection average precision (AP), whereas image dropping achieves only 74.2% of the baseline AP, highlighting the advantage of scrubbing in preserving dataset utility. The degradation is even more severe for small objects due to occlusion and loss of fine-grained details. Furthermore, in NeRF-based 3D reconstruction, our method incurs a PSNR loss of at most 1.66 dB while maintaining SSIM and improving LPIPS, demonstrating superior perceptual quality. Our findings establish object removal as an effective privacy framework, achieving strong privacy guarantees with minimal performance trade-offs. The results highlight key challenges in generative inpainting, occlusion-robust segmentation, and task-specific scrubbing, setting the foundation for future advancements in privacy-preserving vision systems.