Intelligent Multi-View Test Time Augmentation

Personal use of this material is permitted. ABSTRACT In this study, we introduce an intelligent Test Time Augmentation (TTA) algorithm designed to enhance the robustness and accuracy of image classification models against viewpoint variations. Unlike traditional TTA methods that indiscriminately apply augmentations, our approach intelligently selects optimal augmentations based on predictive uncertainty Figure 1: Comparison of Intelligent Multi-View TTA with the metrics. This selection is achieved via a two-stage process: conventional single-view method. This illustrates how the intelligent the first stage identifies the optimal augmentation for each approach dynamically selects augmentation views to class by evaluating uncertainty levels, while the second stage refine predictions (P), in contrast to the conventional method's implements an uncertainty threshold to determine when applying reliance on a single, static view. This methodological advancement ensures that augmentations contribute to classification more effectively than a uniform application across the to the inference phase, applying augmentations to test data dataset.