Research map / Vision-language and multimodal / Cross-modal learning
SARA: Structure-Aware Riemannian-Guided Alignment for Drone Image-Text Retrieval
Keyu Lu, Qing Ma, Zhenyu Lu, Cong Bai
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 visual structural features (VSF) from SPD covariance (ablation) | semantic baseline · ERA DIT | mR: 44.11 -> 45.12 | Improved |
| add visual structural features (VSF) from SPD covariance (ablation) | semantic baseline · UDV DIT | mR: 21.02 -> 21.35 | Improved |
| add textual structural features (TSF) alone (ablation) | semantic baseline · ERA DIT | mR: 44.11 -> 36.86 | Hurt |
| add textual structural features (TSF) alone (ablation) | semantic baseline · UDV DIT | mR: 21.02 -> 20.26 | Hurt |
| add TSF on top of VSF (ablation) | MSFE · ERA DIT | mR: 45.12 -> 45.43 | Improved |
| add marginal distribution alignment (MDA) (ablation) | MSFE (VSF+TSF) · ERA DIT | mR: 45.43 -> 46.62 | Improved |
| add marginal distribution alignment (MDA) (ablation) | MSFE (VSF+TSF) · UDV DIT | mR: 21.57 -> 21.79 | Improved |
| add conditional distribution alignment (CDA) (ablation) | MSFE (VSF+TSF) · ERA DIT | mR: 45.43 -> 46.81 | Improved |
| add CDA on top of MDA (full LESA) (ablation) | SARA · ERA DIT | mR: 46.62 -> 48.36 | Improved |
| add MDA and CDA without structural features (ablation) | semantic baseline · ERA DIT | mR: 44.11 -> 45.50 | Improved |
| set structural loss weight alpha to 0.7 instead of 1.0 (ablation) | SARA · ERA DIT | mR: 43.52 -> 48.36 | Improved |
| set structural loss weight alpha to 0.7 instead of 0.01 (ablation) | SARA · ERA DIT | mR: 45.23 -> 48.36 | Improved |
Closest papers by meaning
- Learning Fine-Grained Alignment for Aerial Vision-Dialog Navigation (AAAI 2025 · similarity 0.86)
- Topology-Aware Representation Alignment for Semi-Supervised Vision-Language Learning (NeurIPS 2026 · similarity 0.84)
- Escaping Plato's Cave: Towards the Alignment of 3D and Text Latent Spaces (CVPR 2025 · similarity 0.82)
- Tango3D: Towards Alignment for Global and Local 2D-3D Correspondence (NeurIPS 2026 · similarity 0.82)
- Weakly Misalignment-free Adaptive Feature Alignment for UAVs-based Multimodal Object Detection (CVPR 2024 · similarity 0.82)
- Object-level Geometric Structure Preserving for Natural Image Stitching (AAAI 2025 · similarity 0.81)
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.
