PartImageNet++ Dataset: Scaling up Part-based Models for Robust Recognition
This work addresses the robustness issue in object recognition for AI systems, but it is incremental as it builds on existing part-based models by scaling up with a new dataset.
The authors tackled the problem of weak adversarial robustness in deep learning-based object recognition by introducing PartImageNet++, a dataset with part segmentation annotations for ImageNet-1K, and a Multi-scale Part-supervised Model (MPM) that improved robustness over strong baselines across various attacks, common corruptions, and out-of-distribution datasets.
Deep learning-based object recognition systems can be easily fooled by various adversarial perturbations. One reason for the weak robustness may be that they do not have part-based inductive bias like the human recognition process. Motivated by this, several part-based recognition models have been proposed to improve the adversarial robustness of recognition. However, due to the lack of part annotations, the effectiveness of these methods is only validated on small-scale nonstandard datasets. In this work, we propose PIN++, short for PartImageNet++, a dataset providing high-quality part segmentation annotations for all categories of ImageNet-1K (IN-1K). With these annotations, we build part-based methods directly on the standard IN-1K dataset for robust recognition. Different from previous two-stage part-based models, we propose a Multi-scale Part-supervised Model (MPM), to learn a robust representation with part annotations. Experiments show that MPM yielded better adversarial robustness on the large-scale IN-1K over strong baselines across various attack settings. Furthermore, MPM achieved improved robustness on common corruptions and several out-of-distribution datasets. The dataset, together with these results, enables and encourages researchers to explore the potential of part-based models in more real applications.