AI Paper Atlas

Research map / Detection and segmentation / Tracking

FAMNet: Joint Learning of Feature, Affinity and Multi-Dimensional Assignment for Online Multiple Object Tracking

Peng Chu, Haibin Ling

ICCV 2019 · Detection and segmentation › Tracking · approach: Association cues for online MOT · 248 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
fine-tune from ILSVRC15 pretrained weights instead of training from scratch (ablation)FAMNet · MOT2015 ablation sequencesMOTA: 44.1 -> 45.2Improved
train the network on MOT sequence instead of random initialization (ablation)FAMNet · MOT2015 ablation sequencesMOTA: 35.5 -> 44.1Improved
add dedicated CNN_BBE bounding box quality estimator (ablation)FAMNet · MOT2015 ablation sequencesMOTA: 40.5 -> 45.2Improved
add CNN_BBE bounding box quality estimator (ablation)FAMNet · MOT2015 ablation sequencesFP: 518 -> 259Improved
add SOT predictions via virtual candidates (ablation)FAMNet · MOT2015 ablation sequencesMOTA: 42.0 -> 45.2Improved
add SOT predictions via virtual candidates (ablation)FAMNet · MOT2015 ablation sequencesFN: 4412 -> 4105Improved
add SOT predictions via virtual candidates (ablation)FAMNet · MOT2015 ablation sequencesFP: 200 -> 259Hurt
add SOT predictions via virtual candidates (ablation)FAMNet · MOT2015 ablation sequencesIDS: 99 -> 87Improved
use end-to-end learned higher-order affinity instead of RNN end-to-end online trackersFAMNet · MOT2015 testMOTA: 19.0 -> 40.6Improved
use FAMNet instead of hand-crafted R1TAFAMNet · KITTI-Car testMOTA: 71.2 -> 77.1Improved
use FAMNet instead of best online baselineFAMNet · MOT2017 testMOTA: 50.9 -> 52.0Improved
use FAMNet instead of best offline methodFAMNet · KITTI-Car testMOTA: 78.1 -> 77.1Hurt

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