Structured-Noise Masked Modeling for Video, Audio and Beyond
This work addresses the need for modality-aware masking strategies in self-supervised learning for video and audio representation learning, offering an incremental improvement over existing methods.
The paper tackles the problem of random masking in self-supervised learning by introducing structured noise-based masking that aligns with spatial, temporal, and spectral characteristics of video and audio data, resulting in consistent performance improvements over random masking without computational overhead.
Masked modeling has emerged as a powerful self-supervised learning framework, but existing methods largely rely on random masking, disregarding the structural properties of different modalities. In this work, we introduce structured noise-based masking, a simple yet effective approach that naturally aligns with the spatial, temporal, and spectral characteristics of video and audio data. By filtering white noise into distinct color noise distributions, we generate structured masks that preserve modality-specific patterns without requiring handcrafted heuristics or access to the data. Our approach improves the performance of masked video and audio modeling frameworks without any computational overhead. Extensive experiments demonstrate that structured noise masking achieves consistent improvement over random masking for standard and advanced masked modeling methods, highlighting the importance of modality-aware masking strategies for representation learning.