Research map / Video / Video segmentation and tracking
Meta Spatio-Temporal Debiasing for Video Scene Graph Generation
Li Xu, Haoxuan Qu, Jason Kuen, Jiuxiang Gu, Jun Liu
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 spatial conditional-bias query sets (ablation) | VidVRD-II · VidVRD | mAP: 30.49 -> 31.57 | Improved |
| add temporal conditional-bias query sets (ablation) | VidVRD-II · VidVRD | mAP: 30.47 -> 31.57 | Improved |
| add predicate-centered spatial biases (ablation) | VidVRD-II · VidVRD | mAP: 30.98 -> 31.57 | Improved |
| add subject-centered spatial biases (ablation) | VidVRD-II · VidVRD | mAP: 31.10 -> 31.57 | Improved |
| add object-centered spatial biases (ablation) | VidVRD-II · VidVRD | mAP: 31.05 -> 31.57 | Improved |
| add forward-case temporal biases (ablation) | VidVRD-II · VidVRD | mAP: 30.94 -> 31.57 | Improved |
| add backward-case temporal biases (ablation) | VidVRD-II · VidVRD | mAP: 31.00 -> 31.57 | Improved |
| apply the meta-debiasing framework to VidVRD-II | VidVRD-II · VidVRD | mAP: 29.37 -> 31.57 | Improved |
| apply the meta-debiasing framework to VRD-STGC | VRD-STGC · VidVRD | mAP: 18.38 -> 20.76 | Improved |
| apply the meta-debiasing framework to the independent baseline | Independent baseline · VidVRD | mAP: 27.49 -> 30.02 | Improved |
| apply the framework with spatial query sets to VCTree on image SGG | VCTree · Visual Genome PredCls | mR@100: 19.1 -> 31.4 | Improved |
| apply the framework to BGNN on image SGG | BGNN · Visual Genome PredCls | mR@100: 32.9 -> 34.1 | Improved |
Closest papers by meaning
- Towards Unbiased and Robust Spatio-Temporal Scene Graph Generation and Anticipation (CVPR 2025 · similarity 0.90)
- TD²-Net: Toward Denoising and Debiasing for Video Scene Graph Generation (AAAI 2024 · similarity 0.89)
- DiffVsgg: Diffusion-Driven Online Video Scene Graph Generation (CVPR 2025 · similarity 0.88)
- Weakly Supervised Video Scene Graph Generation via Natural Language Supervision (ICLR 2025 · similarity 0.88)
- A Closer Look at Dynamic Scene Graph Generation in the Era of Multimodal Large Language Models (NeurIPS 2026 · similarity 0.87)
- OED: Towards One-stage End-to-End Dynamic Scene Graph Generation (CVPR 2024 · 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.
