Research map / Graph learning / Knowledge and temporal graphs
CIGAR: Cross-Modality Graph Reasoning for Domain Adaptive Object Detection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use CIGAR instead of SIGMA | UDA object detection with VGG-16 · Sim10k→Cityscapes | mAP: 53.7 -> 58.5 | Helped |
| Use CIGAR instead of SIGMA | UDA object detection with VGG-16 · KITTI→Cityscapes | mAP: 45.8 -> 48.5 | Helped |
| Use CIGAR instead of SIGMA | UDA object detection with VGG-16 · Cityscapes→Foggy Cityscapes | mAP: 43.5 -> 44.7 | Helped |
| Use CIGAR instead of SIGMA | UDA object detection with ResNet-50 · Cityscapes→Foggy Cityscapes | mAP: 44.2 -> 44.9 | Helped |
| Use CIGAR instead of UMT | UDA object detection with ResNet-101 · Pascal VOC→Clipart | mAP: 44.1 -> 46.2 | Helped |
| Select visual graph nodes by LSV-based DFS instead of random selection (ablation) | CIGAR · Sim10k→Cityscapes | mAP0.5: 56.3 -> 58.5 | Helped |
| Select visual graph nodes by LSV-based DFS instead of uniform selection (ablation) | CIGAR · Sim10k→Cityscapes | mAP0.5: 55.6 -> 58.5 | Helped |
| Add visual intra-graph reasoning (ablation) | CIGAR · Sim10k→Cityscapes | mAP: 49.9 -> 51.4 | Helped |
| Add linguistic intra-graph reasoning (ablation) | CIGAR · Cityscapes→Foggy Cityscapes | mAP: 42.8 -> 42.2 | Hurt |
| Add cross-modality graph reasoning (ablation) | CIGAR · Sim10k→Cityscapes | mAP: 49.9 -> 51.3 | Helped |
| Use all three graph reasoning modules rather than any reported partial combination (ablation) | CIGAR · Sim10k→Cityscapes | mAP: 56.6 -> 58.5 | Helped |
| Add linguistic graph reasoning to the visual-graph baseline (ablation) | CIGAR · Sim10k→Cityscapes | mAP: 54.5 -> 57.6 | Helped |
Closest papers by meaning
- SIGMA: Semantic-Complete Graph Matching for Domain Adaptive Object Detection (CVPR 2022 · similarity 0.90)
- Mario: Multimodal Graph Reasoning with Large Language Models (CVPR 2026 · similarity 0.86)
- SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection (CVPR 2025 · similarity 0.86)
- CRIT: Graph-Based Automatic Data Synthesis to Enhance Cross-Modal Multi-Hop Reasoning (CVPR 2026 · similarity 0.85)
- DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object Detection (NeurIPS 2024 · similarity 0.85)
- Unified Domain Generalization and Adaptation for Multi-View 3D Object Detection (NeurIPS 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.