AI Paper Atlas

Research map / 3D vision / Gaussian splatting

DefenseSplat: Enhancing the Robustness of 3D Gaussian Splatting via Frequency-Aware Filtering

Yiran Qiao, Yiren Lu, Yunlai Zhou, Rui Yang, Linlin Hou, Yu Yin, Jing Ma

ECCV 2026 · 3D vision › Gaussian splatting · approach: Regularization and primitive design for 3DGS · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ReLU scale regularization loss (ablation)DefenseSplat · Mip-NeRF 360 (attacked, eps 16/255)PSNR: 27.32 -> 27.47Improved
add ReLU scale regularization loss (ablation)DefenseSplat · Mip-NeRF 360 (attacked, eps 16/255)GPU memory (MB): 12567 -> 11400Improved
add ReLU scale regularization loss (ablation)DefenseSplat · Tanks-and-Temples (8 scenes, attacked, eps 16/255)PSNR: 26.29 -> 26.42Improved
add ReLU scale regularization loss (ablation)DefenseSplat · LLFF (attacked, eps 16/255)PSNR: 27.67 -> 27.63No clear effect
add ReLU scale loss under strong attack (ablation)DefenseSplat · Tanks-and-Temples (attacked, eps 64/255)PSNR: 19.27 -> 19.38Improved
apply DefenseSplat to attacked inputs instead of plain 3DGS3DGS · Mip-NeRF 360 (attacked, eps 16/255)PSNR: 25.18 -> 27.32Improved
apply DefenseSplat to attacked inputs instead of plain 3DGS3DGS · Mip-NeRF 360 (attacked, eps 16/255)GPU memory (MB): 20965 -> 12567Improved
increase attack strength to 64/255 with DefenseSplatDefenseSplat · Tanks-and-TemplesPSNR: 24.89 -> 19.27Hurt
apply DefenseSplat to clean input3DGS · Mip-NeRF 360 (clean input)PSNR: 29.06 -> 28.58Hurt
apply DefenseSplat to clean input3DGS · LLFF (clean input)PSNR: 31.29 -> 28.39Hurt
apply DefenseSplat to clean input3DGS · Mip-NeRF 360 (clean input)GPU memory (MB): 20965 -> 11012Improved
add fine-grained cross-view consistency filter (ablation)DefenseSplat + ReLU · Mip-NeRF 360 (basic vs fine-grained filter)PSNR: 27.47 -> 27.49No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy