AI Paper Atlas

Research map / Detection and segmentation / Tracking

Language Decoupling with Fine-grained Knowledge Guidance for Referring Multi-object Tracking

Guangyao Li, Siping Zhuang, Yajun Jian, Yan Yan, Hanzi Wang

ICCV 2025 · Detection and segmentation › Tracking · approach: Association cues for online MOT · 3 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 MPA module to base (ablation)DKGTrack · Refer-KITTI valHOTA: 50.31 -> 51.07Improved
add SSE module to base (ablation)DKGTrack · Refer-KITTI valHOTA: 50.31 -> 51.36Improved
add SSE module to base (ablation)DKGTrack · Refer-KITTI valDetA: 38.49 -> 40.53Improved
add SSE module to base (ablation)DKGTrack · Refer-KITTI valAssA: 65.93 -> 65.58Hurt
add SSE on top of MPA (ablation)DKGTrack · Refer-KITTI valHOTA: 51.07 -> 52.08Improved
add SSE on top of MPA (ablation)DKGTrack · Refer-KITTI valAssA: 66.20 -> 66.04Hurt
initialize queries with sentence embedding instead of word-level features (ablation)DKGTrack · Refer-KITTI valHOTA: 50.51 -> 51.36Improved
initialize queries with sentence embedding instead of localized descriptions only (ablation)DKGTrack · Refer-KITTI valHOTA: 50.87 -> 51.36Improved
set temperature beta to 0.5 instead of 0.1 (ablation)DKGTrack MPA · Refer-KITTI valHOTA: 50.33 -> 51.07Improved
set temperature beta to 0.5 instead of 0.9 (ablation)DKGTrack MPA · Refer-KITTI valHOTA: 50.21 -> 51.07Improved
use DKGTrack instead of TempRMOTDKGTrack · Refer-KITTI testHOTA: 50.31 -> 52.08Improved
use DKGTrack instead of TempRMOTDKGTrack · Refer-KITTI-V2 testHOTA: 33.79 -> 35.26Improved

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