Nima Sedaghat, Mohammadreza Zolfaghari, Ehsan Amiri, Thomas Brox
Recent work has shown good recognition results in 3D object recognition using 3D convolutional networks. In this paper, we show that the object orientation plays an important role in 3D recognition. More specifically, we argue that objects induce different features in the network under rotation. Thus, we approach the category-level classification task as a multi-task problem, in which the network is trained to predict the pose of the object in addition to the class label as a parallel task. We show that this yields significant improvements in the classification results. We test our suggested architecture on several datasets representing various 3D data sources: LiDAR data, CAD models, and RGB-D images. We report state-of-the-art results on classification as well as significant improvements in precision and speed over the baseline on 3D detection.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| 3D | ModelNet10 | Accuracy | 93.8 | ORION |
| Shape Representation Of 3D Point Clouds | Sydney Urban Objects | F1 | 77.8 | ORION |
| Shape Representation Of 3D Point Clouds | ModelNet10 | Accuracy | 93.8 | ORION |
| 3D Object Classification | ModelNet10 | Accuracy | 93.8 | ORION |
| 3D Point Cloud Classification | Sydney Urban Objects | F1 | 77.8 | ORION |
| 3D Point Cloud Classification | ModelNet10 | Accuracy | 93.8 | ORION |
| 3D Classification | ModelNet10 | Accuracy | 93.8 | ORION |
| 3D Point Cloud Reconstruction | Sydney Urban Objects | F1 | 77.8 | ORION |
| 3D Point Cloud Reconstruction | ModelNet10 | Accuracy | 93.8 | ORION |