Research map / Detection and segmentation / 3D detection and driving
Trust, but Verify: Cross-Modality Fusion for HD Map Change Detection
John Lambert, James Hays
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 |
|---|---|---|---|
| replace late fusion with early fusion (ego-view) (ablation) | map change detector, ResNet-18 · TbV-Beta test | mAcc: 0.4930 -> 0.6724 | Improved |
| replace late fusion with early fusion (BEV) (ablation) | map change detector, ResNet-50 · TbV-Beta test | mAcc: 0.5761 -> 0.6728 | Improved |
| use ego-view instead of bird's eye view (ablation) | map change detector · TbV-Beta test | mAcc: 5% more effective | Improved |
| use ResNet-18 instead of ResNet-50 in ego-view (ablation) | early fusion ego-view · TbV-Beta test | mAcc | Improved |
| add semantic label maps to sensor+map input (ego-view) (ablation) | early fusion ResNet-18 · TbV-Beta test | mAcc: 0.6463 -> 0.6724 | Improved |
| add semantic label maps to sensor+map input (BEV) (ablation) | early fusion ResNet-50 · TbV-Beta test | mAcc: 0.6543 -> 0.6728 | Improved |
| remove sensor data (map-only input), ego-view (ablation) | early fusion ResNet-18 · TbV-Beta test | mAcc: 0.6724 -> 0.5333 | Hurt |
| remove sensor data (map-only input), BEV (ablation) | early fusion ResNet-50 · TbV-Beta test | mAcc: 0.6728 -> 0.5754 | Hurt |
| random dropout of semantics or map stream (50% per example) (ablation) | early fusion ego-view ResNet-18 · TbV-Beta test | mAcc: 0.6724 -> 0.6850 | Improved |
| dropout on RGB and semantics streams (ablation) | early fusion ego-view ResNet-18 · TbV-Beta test | mAcc: 0.6724 -> 0.6581 | Hurt |
| dropout on semantics stream only (ablation) | early fusion ego-view ResNet-18 · TbV-Beta test | mAcc: 0.6724 -> 0.6483 | Hurt |
| dropout on map stream only (ablation) | early fusion ego-view ResNet-18 · TbV-Beta test | mAcc: 0.6724 -> 0.6617 | Hurt |
Closest papers by meaning
- SDTagNet: Leveraging Text-Annotated Navigation Maps for Online HD Map Construction (NeurIPS 2025 · similarity 0.85)
- The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generationf (CVPR 2025 · similarity 0.85)
- Is Your HD Map Constructor Reliable under Sensor Corruptions? (NeurIPS 2024 · similarity 0.85)
- MapExpert: Online HD Map Construction with Simple and Efficient Sparse Map Element Expert (AAAI 2025 · similarity 0.85)
- Unveiling the Hidden: Online Vectorized HD Map Construction with Clip-Level Token Interaction and Propagation (NeurIPS 2024 · similarity 0.84)
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · 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.
