AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Pair Then Relation: Pair-Net for Panoptic Scene Graph Generation

Jinghao Wang, Zhengyu Wen, Xiangtai Li, Zujin Guo, Jingkang Yang, Ziwei Liu

TPAMI 2024 · Graph learning › Knowledge and temporal graphs · approach: Scene graph generation · 38 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
use Pair-Net pair-then-relation design instead of PSGFormer+PSG · PSG valR@20: 18.9 -> 29.6Improved
use Pair-Net instead of PSGFormer+PSG · PSG valPair R@20: 28.6 -> 52.7Improved
use Swin-B backbone instead of ResNet-50Pair-Net · PSG valR@20: 29.6 -> 33.3Improved
add BCE supervision of the pair matrix (ablation)Pair Proposal Network · PSG valmR@20: 0.5 -> 24.7Improved
add Matrix Learner (ablation)Pair Proposal Network · PSG valmR@20: 14.8 -> 24.7Improved
add linear embedding for subject and object (ablation)Pair Proposal Network · PSG valmR@20: 14.6 -> 24.7Improved
use CNN-tiny matrix learner instead of MLP (ablation)Matrix Learner · PSG valmR@20: 13.0 -> 24.7Improved
scale matrix learner from CNN-tiny to CNN-base (0.2M to 30M) (ablation)Matrix Learner · PSG valmR@20 / R@20: 24.7 / 29.6 -> 23.3 / 33.3Mixed
use Seesaw loss instead of cross entropy (ablation)relation classification · PSG valmR@20: 15.4 -> 24.7Improved
use concat subject-object pairs instead of image features as KV (ablation)Relation Fusion · PSG valmR@20: 1.3 -> 24.7Improved
change number of relation queries from 100 to 200 (ablation)Relation Fusion · PSG valmR@20: 24.7 -> 22.3No clear effect
increase pair loss weight from 5 to 10 (ablation)Pair-Net · PSG valmR@20: 24.7 -> 22.0No clear effect

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