Research map / Probabilistic and scientific modeling / Molecules and proteins
Towards Safer and Understandable Driver Intention Prediction
Mukilan Karuppasamy, Shankar Gangisetty, Shyam Nandan Rai, Carlo Masone, C V Jawahar
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 LTM and LCBM to I3D with bottleneck | I3D wB · DAAD-X | explanation accuracy: 25.26 -> 28.31 | Improved |
| add LTM to MViTv2 with bottleneck | MViTv2 wB · DAAD-X | explanation accuracy: 25.35 -> 31.22 | Improved |
| add LTM to MViTv2 with bottleneck | MViTv2 wB · DAAD-X | action accuracy: 63.29 -> 69.73 | Improved |
| add LTM to I3D with bottleneck | I3D wB · DAAD-X | action accuracy: 74.09 -> 73.21 | Hurt |
| add a bottleneck layer to I3D | I3D · DAAD-X | action accuracy: 74.78 -> 74.09 | Hurt |
| add a bottleneck layer to VideoMAE | VideoMAE · DAAD-X | action accuracy: 72.5 -> 67.01 | Hurt |
| add LTM to I3D (no LCBM) (ablation) | I3D · DAAD-X component ablation | explanation accuracy: 11.22 -> 26.03 | Improved |
| add LCBM to I3D (no LTM) (ablation) | I3D · DAAD-X component ablation | action accuracy: 68.1 -> 74.09 | Improved |
| add LTM on top of LCBM (ablation) | I3D · DAAD-X component ablation | action accuracy: 74.09 -> 73.21 | Hurt |
| add LTM on top of LCBM (ablation) | I3D · DAAD-X component ablation | explanation accuracy: 25.26 -> 28.31 | Improved |
| use full token aggregation instead of CLS token (ablation) | I3D wB · DAAD-X | explanation accuracy: 23.15 -> 28.31 | Improved |
| increase LTM clusters from 5 to 10 (ablation) | I3D + LTM wB · DAAD-X | explanation accuracy: 28.31 -> 23.92 | Hurt |
Closest papers by meaning
- Early Anticipation of Driving Maneuvers (ECCV 2024 · similarity 0.90)
- InAction: Interpretable Action Decision Making for Autonomous Driving (ECCV 2022 · similarity 0.90)
- Advisable Learning for Self-Driving Vehicles by Internalizing Observation-to-Action Rules (CVPR 2020 · similarity 0.89)
- Where, What, Why: Towards Explainable Driver Attention Prediction (ICCV 2025 · similarity 0.88)
- Continuously Learning, Adapting, and Improving: A Dual-Process Approach to Autonomous Driving (NeurIPS 2024 · similarity 0.87)
- PSI: A Benchmark for Human Interpretation and Response in Traffic Interactions (NeurIPS 2025 · similarity 0.87)
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.
