Research map / Vision-language and multimodal / Cross-modal learning
AiSCREAM: Absolute Target Localization with Language-Conditioned Cross-View Alignment for Autonomous Vehicles
Kei Katsumata, Jun Piao, Naoki Hosomi, Kentaro Yamada, Komei Sugiura
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 MLLM Reshape Localizer refinement (ablation) | AiSCREAM · GRiN-Drive | RMSE (m): 4.95 -> 4.72 | Improved |
| add MLLM Reshape Localizer refinement (ablation) | AiSCREAM · DRAMATiST | RMSE (m): 5.19 -> 5.00 | Improved |
| add aerial semantic (AiS) view (ablation) | AiSCREAM · GRiN-Drive | RMSE (m): 6.60 -> 4.95 | Improved |
| add aerial semantic (AiS) view (ablation) | AiSCREAM · DRAMATiST | RMSE (m): 5.57 -> 5.19 | Improved |
| add cross-view consistent semantic segmentation (CVSS) (ablation) | AiSCREAM · GRiN-Drive | RMSE (m): 5.80 -> 4.95 | Improved |
| add AiS-view segmentation loss term (ablation) | HC loss · GRiN-Drive | RMSE (m): 5.17 -> 4.95 | Improved |
| add obstacle penalty term (ablation) | HC loss · GRiN-Drive | RMSE (m): 5.27 -> 4.95 | Improved |
| add obstacle penalty term (ablation) | HC loss · DRAMATiST | RMSE (m): 5.24 -> 5.19 | No clear effect |
| use AiS view alone instead of front view alone (ablation) | AiSCREAM · GRiN-Drive | RMSE (m): 6.60 -> 7.70 | Hurt |
| use both front and AiS views with alignment (ablation) | AiSCREAM · GRiN-Drive | RMSE (m): 6.60 -> 4.95 | Improved |
| replace BEV representation with AiS view (ablation) | AiSCREAM · GRiN-Drive | RMSE (m): 6.48 -> 4.72 | Improved |
| AiSCREAM versus best baseline | AiSCREAM · DRAMATiST | RMSE (m): 8.56 -> 5.00 | Improved |
Closest papers by meaning
- Think Before You Drive: World Model-Inspired Multimodal Grounding (CVPR 2026 · similarity 0.84)
- LangMap: A Human-Verified Benchmark for Hierarchical Open-Vocabulary Goal Navigation (NeurIPS 2026 · similarity 0.84)
- SpaceDrive: Infusing Spatial Awareness into VLM-based Autonomous Driving (CVPR 2026 · similarity 0.84)
- Prune2Drive: A Plug-and-Play Framework for Accelerating Vision-Language Models in Autonomous Driving (CVPR 2026 · similarity 0.84)
- SURDS: Benchmarking Spatial Understanding and Reasoning in Driving Scenarios with Vision Language Models (NeurIPS 2025 · similarity 0.83)
- BeliefMapNav: 3D Voxel-Based Belief Map for Zero-Shot Object Navigation (NeurIPS 2025 · similarity 0.83)
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.
