Research map / Graph learning / Knowledge and temporal graphs
From Easy to Hard: Learning Curricular Shape-Aware Features for Robust Panoptic Scene Graph Generation
Hanrong Shi, Lin Li, Jun Xiao, Yueting Zhuang, Long Chen
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add CAFE | Motifs · PSG PredCls | Mean: 35.5 -> 39.7 | Improved |
| add CAFE | VCTree · PSG PredCls | Mean: 35.9 -> 40.6 | Improved |
| add CAFE | Transformer · PSG PredCls | Mean: 27.9 -> 40.0 | Improved |
| add CAFE | Motifs · PSG SGDet | Mean: 15.3 -> 25.7 | Improved |
| add CAFE | Motifs · PSG PredCls | R@100: 52.4 -> 47.2 | Mixed |
| add all three features (bbox, mask, boundary) over bbox only (ablation) | Motifs+CAFE · PSG PredCls | Mean: 35.9 -> 39.7 | Improved |
| add curriculum learning over single boundary feature (ablation) | Motifs+CAFE · PSG PredCls | Mean: 36.5 -> 39.7 | Improved |
| use cognition-based predicate grouping instead of random (ablation) | Motifs+CAFE · PSG PredCls | Mean: 37.7 -> 39.7 | Improved |
| use cognition-based predicate grouping instead of average (ablation) | Motifs+CAFE · PSG PredCls | Mean: 38.5 -> 39.7 | Improved |
| add oversampling and median resampling (ablation) | Motifs+CAFE · PSG PredCls | Mean: 37.1 -> 39.7 | Improved |
| entangled attention instead of concatenation fusion (ablation) | Transformer+CAFE · PSG PredCls | Mean: 35.2 -> 40.0 | Improved |
| lambda 0.5 instead of 0 (ablation) | Motifs+CAFE · PSG PredCls | Mean: 35.8 -> 39.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- HiLo — HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation (ICCV 2023) · better on 0 of 1 shared result
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
- Panoptic Scene Graph Generation (ECCV 2022 · similarity 0.91)
- A Fair Ranking and New Model for Panoptic Scene Graph Generation (ECCV 2024 · similarity 0.90)
- OpenPSG: Open-set Panoptic Scene Graph Generation via Large Multimodal Models (ECCV 2024 · similarity 0.89)
- Panoptic Scene Graph Generation with Semantics-Prototype Learning (AAAI 2024 · similarity 0.87)
- DSGG: Dense Relation Transformer for an End-to-end Scene Graph Generation (CVPR 2024 · similarity 0.87)
- Towards Unbiased and Robust Spatio-Temporal Scene Graph Generation and Anticipation (CVPR 2025 · similarity 0.86)
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.
