AI Paper Atlas

Research map / Video / Video segmentation and tracking

Dual-Temporal Exemplar Representation Network for Video Semantic Segmentation

Xiaolong Xu, Lei Zhang, Jiayi Li, Lituan Wang, Yifan Guan, Yu Yan, Leyi Zhang, Hao Song

ICCV 2025 · Video › Video segmentation and tracking · approach: Video semantic and panoptic segmentation · 1 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 Global Temporal Exemplar Module (ablation)DTERN · VSPW val (ablation, MiT-B1)VEC8: 37.6 -> 37.8Improved
add Global Temporal Exemplar Module (ablation)DTERN · VSPW val (ablation, MiT-B1)VEC16: 38.8 -> 39.0Improved
add Global Temporal Exemplar Module (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 40.4 -> 40.3Hurt
add cluster center aggregation (SAM) instead of direct summation (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 40.0 -> 40.3Improved
use CAM instead of ResBlock (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 39.7 -> 40.3Improved
use 200 clusters instead of 32 (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 39.3 -> 40.3Improved
use 200 clusters instead of 124 (K=C) (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 39.7 -> 40.3Improved
use 200 clusters instead of 2048 (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 39.6 -> 40.3Improved
use proposed clustering instead of ClusterFormer (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 38.0 -> 40.3Improved
use proposed clustering instead of PaCaViT (ablation)DTERN · VSPW val (ablation, MiT-B1)mIoU: 39.7 -> 40.3Improved
use VEC instead of VC to measure temporal consistencyCFFM predictions · VSPWPearson correlation with ground-truth TC: 59.4 -> 88.1Improved
DTERN instead of SegFormer baselineDTERN with MiT-B0 · VSPW valmIoU: 33.1 -> 35.9Improved

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