LocalNorm: Robust Image Classification through Dynamically Regularized Normalization
This work addresses robustness in image classification for applications where data quality varies, though it is incremental as it builds on existing normalization techniques.
The paper tackles the problem of neural networks being overly sensitive to image degradation by introducing LocalNorm, a variant of Batch Normalization that dynamically adapts to local image intensity and contrast, resulting in up to three times better accuracy on noise-degraded images while maintaining performance on standard benchmarks.
While modern convolutional neural networks achieve outstanding accuracy on many image classification tasks, they are, compared to humans, much more sensitive to image degradation. Here, we describe a variant of Batch Normalization, LocalNorm, that regularizes the normalization layer in the spirit of Dropout while dynamically adapting to the local image intensity and contrast at test-time. We show that the resulting deep neural networks are much more resistant to noise-induced image degradation, improving accuracy by up to three times, while achieving the same or slightly better accuracy on non-degraded classical benchmarks. In computational terms, LocalNorm adds negligible training cost and little or no cost at inference time, and can be applied to already-trained networks in a straightforward manner.