Interpreting Representation Quality of DNNs for 3D Point Cloud Processing
This work addresses representation quality for researchers in 3D computer vision, but it is incremental as it builds on existing DNN analysis methods.
The paper tackles the problem of evaluating knowledge representation quality in deep neural networks for 3D point cloud processing by proposing methods to disentangle vulnerabilities and metrics for spatial smoothness and complexity, with experiments exposing representation issues and explaining adversarial training utility.
In this paper, we evaluate the quality of knowledge representations encoded in deep neural networks (DNNs) for 3D point cloud processing. We propose a method to disentangle the overall model vulnerability into the sensitivity to the rotation, the translation, the scale, and local 3D structures. Besides, we also propose metrics to evaluate the spatial smoothness of encoding 3D structures, and the representation complexity of the DNN. Based on such analysis, experiments expose representation problems with classic DNNs, and explain the utility of the adversarial training.