AI Paper Atlas

Research map / Detection and segmentation / Tracking

Robust Multi-Object Tracking by Marginal Inference

Yifu Zhang, Chunyu Wang, Xinggang Wang, Wenjun Zeng, Wenyu Liu

ECCV 2022 · Detection and segmentation › Tracking · approach: Memory and correlation-based MOT · 17 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 Kalman filter motion cue to appearance matching (ablation)MTracker · MOT17 trainIDF1: 71.9 -> 73.6Improved
add IoU matching after appearance and Kalman (ablation)MTracker · MOT17 trainIDF1: 73.6 -> 74.8Improved
add IoU matching (ablation)MTracker · MOT17 trainID switches: 777 -> 499Improved
add marginal probability matching (ablation)MTracker · MOT17 trainIDF1: 74.8 -> 75.7Improved
add marginal probability matching (ablation)MTracker · MOT17 trainID switches: 499 -> 449Improved
replace feature distance with marginal probability (FairMOT det and ReID)MTracker · MOT17 trainIDF1: 58.0 -> 58.2Improved
replace feature distance with marginal probability (Scaled-YOLOv4 det, FairMOT ReID)MTracker · MOT17 trainIDF1: 72.5 -> 73.6Improved
replace feature distance with marginal probabilityMTracker · MOT20 train (FairMOT)IDF1: 45.9 -> 46.7Improved
replace feature distance with marginal probabilityMTracker · MOT17 trainMOTA: 71.5 -> 71.6No clear effect
use marginal inference instead of softmax probabilityMTracker · MOT17 trainIDF1: 73.7 -> 75.7Improved
use marginal inference instead of bi-directional softmaxMTracker · MOT17 trainIDF1: 74.7 -> 75.7Improved
use marginal inference instead of bi-directional softmaxMTracker · MOT17 trainID switches: 405 -> 449Hurt

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