MINR: Implicit Neural Representations with Masked Image Modelling
This addresses robustness and generalization issues in self-supervised learning for image representation, though it appears incremental as it builds on existing masked autoencoder methods.
The paper tackled the problem of self-supervised learning methods like masked autoencoders being sensitive to masking strategies and out-of-distribution data, by introducing the MINR framework that combines implicit neural representations with masked image modeling, resulting in outperforming MAE in both in-domain and out-of-distribution scenarios while reducing model complexity.
Self-supervised learning methods like masked autoencoders (MAE) have shown significant promise in learning robust feature representations, particularly in image reconstruction-based pretraining task. However, their performance is often strongly dependent on the masking strategies used during training and can degrade when applied to out-of-distribution data. To address these limitations, we introduce the masked implicit neural representations (MINR) framework that synergizes implicit neural representations with masked image modeling. MINR learns a continuous function to represent images, enabling more robust and generalizable reconstructions irrespective of masking strategies. Our experiments demonstrate that MINR not only outperforms MAE in in-domain scenarios but also in out-of-distribution settings, while reducing model complexity. The versatility of MINR extends to various self-supervised learning applications, confirming its utility as a robust and efficient alternative to existing frameworks.