Research map / Vision-language and multimodal / Cross-modal learning
Selective Sensor Fusion for Neural Visual-Inertial Odometry
Check 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 deterministic soft selective fusion instead of direct concatenation | Neural VIO · KITTI vision degradation | rotation error (degrees): 0.164 -> 0.150 | Helped |
| Use deterministic soft selective fusion instead of direct concatenation | Neural VIO · KITTI all degradation | rotation error (degrees): 0.139 -> 0.134 | Helped |
| Use deterministic soft selective fusion instead of direct concatenation | Neural VIO · KITTI all degradation | translation error (m): 0.148 -> 0.152 | Hurt |
| Use stochastic hard selective fusion instead of direct concatenation | Neural VIO · KITTI vision degradation | translation error (m): 0.175 -> 0.172 | Helped |
| Use stochastic hard selective fusion instead of direct concatenation | Neural VIO · KITTI all degradation | rotation error (degrees): 0.139 -> 0.150 | Hurt |
| Use deterministic soft selective fusion instead of direct concatenation | Neural VIO · EuRoC vision degradation | rotation error (degrees): 0.0696 -> 0.0533 | Helped |
| Use stochastic hard selective fusion instead of direct concatenation | Neural VIO · EuRoC all degradation | rotation error (degrees): 0.0935 -> 0.0823 | Helped |
| Use stochastic hard selective fusion instead of direct concatenation | Neural VIO · PennCOSYVIO all degradation | translation error (m): 0.0407 -> 0.0400 | Helped |
| Use stochastic hard selective fusion instead of direct concatenation | Neural VIO · PennCOSYVIO all degradation | rotation error (degrees): 1.353 -> 1.232 | Helped |
| Add feature masks that explicitly select/reweight sensor features | Neural VIO · KITTI, EuRoC, and PennCOSYVIO degraded data | translation and rotation error | Mixed |
Closest papers by meaning
- Adaptive VIO: Deep Visual-Inertial Odometry with Online Continual Learning (CVPR 2024 · similarity 0.90)
- EqNIO: Subequivariant Neural Inertial Odometry (ICLR 2025 · similarity 0.86)
- From Variance to Veracity: Unbundling and Mitigating Gradient Variance in Differentiable Bundle Adjustment Layers (CVPR 2024 · similarity 0.84)
- Leveraging Consistent Spatio-Temporal Correspondence for Robust Visual Odometry (AAAI 2025 · similarity 0.83)
- LEAP-VO: Long-term Effective Any Point Tracking for Visual Odometry (CVPR 2024 · similarity 0.83)
- Adaptive Fusion of Single-View and Multi-View Depth for Autonomous Driving (CVPR 2024 · 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.