Remove360: Benchmarking Residuals After Object Removal in 3D Gaussian Splatting

2025-08-15

AI Summary

This work addresses semantic remnants—residual object information persisting after removal—in 3D Gaussian Splatting (3DGS), exposing critical privacy and controllability risks in editable 3D scene representations. We identify that existing removal methods often retain semantically detectable traces even when geometric structures are erased. To systematically evaluate this issue, we introduce Remove360, the first benchmark for semantic remnant assessment in 3DGS, comprising diverse indoor/outdoor real-world scenes. Our method employs a contrastive analysis framework leveraging pre- and post-removal RGB images and object masks, quantifying remnant strength via downstream model inference (e.g., black-box classifiers). Experiments reveal that state-of-the-art 3DGS removal techniques exhibit significant, exploitable semantic remnants. We publicly release the Remove360 dataset, evaluation code, and standardized protocols—establishing a new benchmark and actionable pathway for privacy-aware and controllable 3D reconstruction.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Natural Language Processing: Safety and Robustness

Application Category

• Security and Privacy: Data transparency and provenance
• Semantics and Knowledge: Provenance, trust, security and privacy, and ethical issues in managing semantic data
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Understanding what semantic information persists after object removal is critical for privacy-preserving 3D reconstruction and editable scene representations. In this work, we introduce a novel benchmark and evaluation framework to measure semantic residuals, the unintended semantic traces left behind, after object removal in 3D Gaussian Splatting. We conduct experiments across a diverse set of indoor and outdoor scenes, showing that current methods can preserve semantic information despite the absence of visual geometry. We also release Remove360, a dataset of pre/post-removal RGB images and object-level masks captured in real-world environments. While prior datasets have focused on isolated object instances, Remove360 covers a broader and more complex range of indoor and outdoor scenes, enabling evaluation of object removal in the context of full-scene representations. Given ground truth images of a scene before and after object removal, we assess whether we can truly eliminate semantic presence, and if downstream models can still infer what was removed. Our findings reveal critical limitations in current 3D object removal techniques and underscore the need for more robust solutions capable of handling real-world complexity. The evaluation framework is available at github.com/spatial-intelligence-ai/Remove360.git. Data are available at huggingface.co/datasets/simkoc/Remove360.

Problem

• Measure semantic residuals after 3D object removal
• Evaluate privacy risks in 3D scene reconstruction
• Assess downstream model inference on removed objects

Innovation

• Benchmark for semantic residuals in 3D Gaussian Splatting
• Dataset with pre/post-removal RGB images and masks
• Evaluation framework for object removal robustness