AI Paper Atlas

Research map / Video / Video segmentation and tracking

Anomaly Detection in Video Sequence With Appearance-Motion Correspondence

Trong-Nguyen Nguyen, Jean Meunier

ICCV 2019 · Video › Video segmentation and tracking · approach: Weakly supervised video anomaly detection · 490 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
score frames with SSIM on the appearance stream instead of the combined motion-appearance score (ablation)proposed model · Traffic-Trainframe-level AP: 0.490 -> 0.798Improved
score frames with SSIM on the appearance stream instead of the combined score (ablation)proposed model · Traffic-Belleviewframe-level AP: 0.751 -> 0.830Improved
use proposed model instead of future frame predictionFuture frame prediction · CUHK Avenueframe-level AUC: 0.851 -> 0.869Improved
use proposed model instead of future frame predictionFuture frame prediction · UCSD Ped2frame-level AUC: 0.954 -> 0.962Improved
use proposed model instead of AMDN on Entrance false alarmsAMDN · Subway Entrancefalse alarms: 4 -> 18Hurt
use proposed model instead of AMDN on Exit true positivesAMDN · Subway Exittrue positives: 19 -> 17Hurt
use proposed model instead of ALOCC on BelleviewALOCC D(R(X)) · Traffic-Belleviewframe-level AP: 0.805 -> 0.751Hurt
use proposed model instead of ALOCC on TrainALOCC D(R(X)) · Traffic-Trainframe-level AP: 0.237 -> 0.490Improved

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