VaLID: Verification as Late Integration of Detections for LiDAR-Camera Fusion
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909502298652672 |
|---|---|
| author | Vats, Vanshika Nizam, Marzia Binta Davis, James |
| author_facet | Vats, Vanshika Nizam, Marzia Binta Davis, James |
| contents | Vehicle object detection benefits from both LiDAR and camera data, with LiDAR offering superior performance in many scenarios. Fusion of these modalities further enhances accuracy, but existing methods often introduce complexity or dataset-specific dependencies. In our study, we propose a model-adaptive late-fusion method, VaLID, which validates whether each predicted bounding box is acceptable or not. Our method verifies the higher-performing, yet overly optimistic LiDAR model detections using camera detections that are obtained from either specially trained, general, or open-vocabulary models. VaLID uses a lightweight neural verification network trained with a high recall bias to reduce the false predictions made by the LiDAR detector, while still preserving the true ones. Evaluating with multiple combinations of LiDAR and camera detectors on the KITTI dataset, we reduce false positives by an average of 63.9%, thus outperforming the individual detectors on 3D average precision (3DAP). Our approach is model-adaptive and demonstrates state-of-the-art competitive performance even when using generic camera detectors that were not trained specifically for this dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_15529 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | VaLID: Verification as Late Integration of Detections for LiDAR-Camera Fusion Vats, Vanshika Nizam, Marzia Binta Davis, James Computer Vision and Pattern Recognition Vehicle object detection benefits from both LiDAR and camera data, with LiDAR offering superior performance in many scenarios. Fusion of these modalities further enhances accuracy, but existing methods often introduce complexity or dataset-specific dependencies. In our study, we propose a model-adaptive late-fusion method, VaLID, which validates whether each predicted bounding box is acceptable or not. Our method verifies the higher-performing, yet overly optimistic LiDAR model detections using camera detections that are obtained from either specially trained, general, or open-vocabulary models. VaLID uses a lightweight neural verification network trained with a high recall bias to reduce the false predictions made by the LiDAR detector, while still preserving the true ones. Evaluating with multiple combinations of LiDAR and camera detectors on the KITTI dataset, we reduce false positives by an average of 63.9%, thus outperforming the individual detectors on 3D average precision (3DAP). Our approach is model-adaptive and demonstrates state-of-the-art competitive performance even when using generic camera detectors that were not trained specifically for this dataset. |
| title | VaLID: Verification as Late Integration of Detections for LiDAR-Camera Fusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.15529 |