AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Self-Supervised Adversarial Training of Monocular Depth Estimation Against Physical-World Attacks

Zhiyuan Cheng, Cheng Han, James Liang, Qifan Wang, Xiangyu Zhang, Dongfang Liu

TPAMI 2024 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial perturbations beyond images · 30 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
self-supervised L0 adversarial trainingMonodepth2 · L0 1/10 attackABSE (m): 6.08 -> 0.25Improved
self-supervised instead of supervised pseudo-label adversarial trainingMonodepth2 · L0 1/10 attackABSE (m): 1.18 -> 0.25Improved
use contrastive learning baseline for hardeningMonodepth2 · benign depth estimationABSE (m): 2.125 -> 3.218Hurt
self-supervised hardeningMonodepth2 · benign depth estimationABSE (m): 2.125 -> 2.16No clear effect
add contrastive loss to self-supervised training (ablation)Monodepth2 · L0 1/10 attackABSE (m): 0.244 vs 0.24 (all three)No clear effect
add self-supervised loss to Con+Sup (ablation)Monodepth2 · L0 1/10 attackABSE (m): 0.849 -> 0.24Improved
train with ground-truth depth instead of pseudo labels for supervised baseline (ablation)Monodepth2 · L0 1/10 attackABSE (m): 0.94 -> 0.75No clear effect
restrict training distance range to 5-10 m (ablation)L0+SelfSup Monodepth2 · robustness evaluated 5-30 mABSEImproved
randomize viewing angle range in training versus fixed 0 degrees (ablation)L0+SelfSup Monodepth2 · L0 and PGD attacksABSE: PGD 0.2: 3.587 -> 3.470Improved
filter unrealistic object positions from training set (ablation)Monodepth2 · L0 1/10 attackABSE (m): 0.251 -> 0.248No clear effect
use inaccurate half-baseline projection (0.27 m instead of 0.54 m) for view synthesis (ablation)Monodepth2 · KITTI-based trainingdepth outline qualityHurt
self-supervised hardening evaluated on unseen target object (SD White)Monodepth2 · L0 1/10 attackABSE (m): 4.262 -> 1.408Improved

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