Generic Deep Networks with Wavelet Scattering
This work addresses the problem of simplifying deep neural network learning for researchers and practitioners by proposing a method to initialize layers with wavelet filters, though it appears incremental as it builds on existing wavelet transform concepts.
The paper tackles object classification by introducing a two-layer wavelet scattering network that uses wavelet transforms without learning or max pooling, achieving efficient performance on complex datasets like CalTech with structural variability and clutter.
We introduce a two-layer wavelet scattering network, for object classification. This scattering transform computes a spatial wavelet transform on the first layer and a new joint wavelet transform along spatial, angular and scale variables in the second layer. Numerical experiments demonstrate that this two layer convolution network, which involves no learning and no max pooling, performs efficiently on complex image data sets such as CalTech, with structural objects variability and clutter. It opens the possibility to simplify deep neural network learning by initializing the first layers with wavelet filters.