AI Paper Atlas

Research map / Video / Video segmentation and tracking

Privacy-Aware Video Anomaly Detection through Orthogonal Subspace Projection

Lei Wang, Wenxiang Diao, Andrew Busch, Jun Zhou, Yongsheng Gao

ICML 2026 · Video › Video segmentation and tracking · approach: Weakly supervised video anomaly detection · 0 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 G-OPL/OPL layers to MGFNMGFN · ShanghaiTech (main)frame-level AUC (%): 75.3 -> 83.7Improved
add G-OPL/OPL layers to RTFMRTFM · UCF-Crime (main)frame-level AUC (%): 74.3 -> 78.3Improved
add G-OPL to TEVADTEVAD · ShanghaiTech (main)frame-level AUC (%): 98.1 -> 98.2No clear effect
add G-OPL instead of OPL aloneRTFM · ShanghaiTech (privacy-method comparison)FPD (identity leakage): 0.30 -> 0.22Improved
add input masking (blur) to RTFMRTFM · ShanghaiTech (privacy-method comparison)frame-level AUC (%): 96.8 -> 93.5Hurt
add differential privacy noise to RTFMRTFM · ShanghaiTech (privacy-method comparison)frame-level AUC (%): 96.8 -> 95.8Hurt
add G-OPL to RTFM instead of GRL adversarialRTFM · ShanghaiTech (privacy-method comparison)frame-level AUC (%): 95.9 -> 97.3Improved
increase ranks k_OPL and k_G-OPL from 4 to 128 (ablation)OPL + G-OPL · ShanghaiTech validationAUC (%): 97.3 -> 87.9Hurt
use two G-OPL layers instead of one early layer (ablation)G-OPL · ShanghaiTech (layer count)FPD: 0.28 -> 0.22Improved
use three G-OPL layers instead of one early layer (ablation)G-OPL · ShanghaiTech (layer count)privacy-per-GFLOP: 0.233 -> 0.073Hurt
suppress face attribute with G-OPL (ablation)G-OPL · ShanghaiTech (attribute-wise)AUC (%): 83.8 -> 85.3Improved
suppress gait attribute with G-OPL (ablation)G-OPL · ShanghaiTech (attribute-wise)AUC (%): 83.8 -> 83.2Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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