Granular Ball Guided Masking: Structure-aware Data Augmentation
This addresses the need for robust data augmentation in computer vision to prevent overfitting with limited or shifting data, though it is incremental as it builds on existing mask-based methods.
The paper tackles the problem of data augmentation lacking structural awareness by proposing Granular Ball Guided Masking (GBGM), which adaptively preserves semantically important regions and suppresses redundant areas, resulting in consistent improvements in image classification, masked image reconstruction, and image tampering detection across multiple benchmarks.
Deep learning models have achieved remarkable success in computer vision but still rely heavily on large-scale labeled data and tend to overfit when data is limited or distributions shift. Data augmentation -- particularly mask-based information dropping -- can enhance robustness by forcing models to explore complementary cues; however, existing approaches often lack structural awareness and risk discarding essential semantics. We propose Granular Ball Guided Masking (GBGM), a structure-aware augmentation strategy guided by Granular Ball Computing (GBC). GBGM adaptively preserves semantically rich, structurally important regions while suppressing redundant areas through a coarse-to-fine hierarchical masking process, producing augmentations that are both representative and discriminative. Extensive experiments on multiple benchmarks demonstrate consistent improvements not only in image classification and masked image reconstruction, but also in image tampering detection, validating the effectiveness and generalization of GBGM across both recognition and forensic scenarios. Simple and model-agnostic, GBGM integrates seamlessly into CNNs and Vision Transformers, offering a practical paradigm for structure-aware data augmentation.