Research map / Graph learning / Knowledge and temporal graphs
An End-To-End Network for Generating Social Relationship Graphs
Arushi Goel, Keng Teck Ma, Cheston Tan
Check 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 GRU-based message passing between person and relationship states (ablation) | MN-CNN · PIPA-relation graph dataset SRRec | accuracy: 49.75% -> 53.56% | Improved |
| Add scene context features (ablation) | SRG-GN · PIPA-relation graph dataset SRRec | accuracy: 51.79% -> 53.56% | Improved |
| Add scene context features (ablation) | SRG-GN · PIPA-relation graph dataset SRGGen | accuracy: 20.24% -> 27.64% | Improved |
| Add scene context features (ablation) | SRG-GN · PISC SRRec | mAP: 69.2 -> 71.6 | Improved |
| Use mean pooling instead of max pooling (ablation) | SRG-GN with 2 time steps · PIPA-relation graph dataset SRRec | accuracy: 52.16% -> 53.56% | Improved |
| Use 2 time steps instead of 1 (ablation) | SRG-GN with max pooling · PIPA-relation graph dataset SRRec | accuracy: 50.41% -> 52.16% | Improved |
| Use 3 time steps instead of 2 (ablation) | SRG-GN with max pooling · PIPA-relation graph dataset SRRec | accuracy: 52.16% -> 51.27% | Hurt |
| Use 2 time steps instead of 1 (ablation) | SRG-GN with mean pooling · PIPA-relation graph dataset SRRec | accuracy: 50.89% -> 53.56% | Improved |
| Use 3 time steps instead of 2 (ablation) | SRG-GN with mean pooling · PIPA-relation graph dataset SRRec | accuracy: 53.56% -> 52.08% | Hurt |
| Apply weighted cross-entropy to penalize underrepresented classes | SRG-GN · PISC SRRec | precision | Improved |
Closest papers by meaning
- Graph-Based Social Relation Reasoning (ECCV 2020 · similarity 0.91)
- Graph Structure Learning on User Mobility Data for Social Relationship Inference (AAAI 2023 · similarity 0.86)
- SocialGPT: Prompting LLMs for Social Relation Reasoning via Greedy Segment Optimization (NeurIPS 2024 · similarity 0.86)
- Relation-aware Graph Attention Model with Adaptive Self-adversarial Training (AAAI 2021 · similarity 0.86)
- Visual Social Relationship Recognition (IJCV 2020 · similarity 0.86)
- DSGG: Dense Relation Transformer for an End-to-end Scene Graph Generation (CVPR 2024 · similarity 0.84)
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.
