AI Paper Atlas

Research map / Video / Action recognition

HiEve: A Large-Scale Benchmark for Human-Centric Video Analysis in Complex Events

Weiyao Lin, Huabin Liu, Shizhan Liu, Yuxi Li, Hongkai Xiong, Guo-Jun Qi, Nicu Sebe

IJCV 2023 · Video › Action recognition · approach: Disentangled action recognition · 148 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 action-guided domain alignment module (ADAM) to HRNet (ablation)HRNet · HiEve pose estimationw-AP@avg: 52.78 -> 53.10Improved
add pose refinement module (PRM) on top of ADAM (ablation)action-guided HRNet · HiEve pose estimationw-AP@avg: 53.10 -> 53.92Improved
use spatial refinement (SR) only instead of both refinements (ablation)pose refinement module · HiEve pose estimationw-AP@avg: 53.20 vs 53.92 with SR+CRHurt
use channel refinement (CR) only (ablation)pose refinement module · HiEve pose estimationw-AP@avg: 53.65 vs 53.92 with SR+CRHurt
predict pose-aware motion from all three stage features m1,m2,m3 instead of m1 only (ablation)pose-aware SlowFast · HiEve action recognitionwf-mAP@avg: 12.36 -> 13.16Improved
use m2 only instead of m1 only (ablation)pose-aware SlowFast · HiEve action recognitionwf-mAP@avg: 12.36 -> 12.58Improved
add pose-aware feature prediction to SlowFastSlowFast · HiEve action recognitionwf-mAP@avg: 12.08 -> 13.16Improved
add action-guided pose estimation to PoseFlow trackingPoseFlow · HiEve pose trackingMOTA: 44.17 -> 45.36Improved
pretrain on HiEve before COCO pose estimationHRNet · COCO keypointsAP: 74.4 -> 74.8Improved
pretrain on HiEve before MOT20 trackingMOTDT · MOT20MOTA: 46.4 -> 47.6Improved
use ground-truth boxes (oracle) for tracking (ablation)IOUTracker · HiEve trackingMOTA: 38.59 -> 97.70Improved
use ground-truth boxes (oracle) for pose estimation (ablation)DHRN · HiEve pose estimationw-AP@avg: 52.78 -> 53.34Improved

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