Research map / Architectures and efficiency / Attention and transformers
Exploring Relational Context for Multi-Task Dense Prediction
David Brüggemann, Menelaos Kanakis, Anton Obukhov, Stamatios Georgoulis, Luc Van Gool
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 automated per-pair context type selection (ATRC) over the best single context type | relational context · NYUD-v2 (HRNet18) | Delta_m: 1.33 -> 1.56 | Improved |
| use T-label context instead of local context | relational context · NYUD-v2 (HRNet18) | Delta_m: -1.34 -> 1.33 | Improved |
| add relational context distillation (ATRC) to multi-task baseline | multi-task network · NYUD-v2 (HRNet18) | Delta_m: -1.89 -> 1.56 | Improved |
| add FPM to ATRC | ATRC · NYUD-v2 (HRNet18) | Delta_m: 1.56 -> 3.57 | Improved |
| add ATRC with FPM vs MTI-Net | ATRC · NYUD-v2 (HRNet48) | Delta_m: 0.15 -> 0.49 | Improved |
| add ATRC with FPM vs MTI-Net | ATRC · PASCAL-Context (HRNet18) | Delta_m: -2.12 -> -1.98 | Improved |
| add ATRC to multi-task baseline | multi-task network · PASCAL-Context (HRNet18) | Delta_m: -6.77 -> -4.45 | Improved |
| add ATRC to ASPP network | ASPP · PASCAL-Context (ResNet-50) | Delta_m: 1.77 -> 2.13 | Improved |
| add ATRC to branched BMTAS backbone | BMTAS · PASCAL-Context (ResNet-50) | Delta_m: -0.55 -> 4.53 | Improved |
| add self-attention blocks to ATRC (ablation) | ATRC · NYUD-v2 (HRNet18) | Delta_m: 0.87 -> 1.56 | Improved |
| use ground-truth T-label region maps instead of predicted (ablation) | T-label context · NYUD-v2 (HRNet18) | Delta_m: 1.33 -> 12.67 | Improved |
| use ground-truth S-label region maps instead of predicted (ablation) | S-label context · NYUD-v2 (HRNet18) | Delta_m: 1.07 -> 14.55 | Improved |
Closest papers by meaning
- Multi-Task Learning for Dense Prediction Tasks: A Survey (TPAMI 2021 · similarity 0.88)
- Learning Multiple Dense Prediction Tasks From Partially Annotated Data (CVPR 2022 · similarity 0.88)
- Unlocking the Potential of Vision Foundation Models for Multi-Task Dense Prediction via Mixture of Task Adapters (TPAMI 2026 · similarity 0.88)
- TaskPrompter: Spatial-Channel Multi-Task Prompting for Dense Scene Understanding (ICLR 2023 · similarity 0.87)
- Going Beyond Multi-Task Dense Prediction with Synergy Embedding Models (CVPR 2024 · similarity 0.87)
- Multi-Task Dense Prediction via Mixture of Low-Rank Experts (CVPR 2024 · similarity 0.85)
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.
