Discrete Wavelet Transform as a Facilitator for Expressive Latent Space Representation in Variational Autoencoders in Satellite Imagery
This addresses the need for better latent representations in remote sensing applications, but it is incremental as it builds on existing VAE and LDM frameworks.
The paper tackles the problem of improving latent space representation in Variational Autoencoders for satellite imagery by proposing ExpDWT-VAE, which integrates Discrete Wavelet Transform to enhance spatial-frequency features, resulting in improved performance across several metrics on a new dataset.
Latent Diffusion Models (LDM), a subclass of diffusion models, mitigate the computational complexity of pixel-space diffusion by operating within a compressed latent space constructed by Variational Autoencoders (VAEs), demonstrating significant advantages in Remote Sensing (RS) applications. Though numerous studies enhancing LDMs have been conducted, investigations explicitly targeting improvements within the intrinsic latent space remain scarce. This paper proposes an innovative perspective, utilizing the Discrete Wavelet Transform (DWT) to enhance the VAE's latent space representation, designed for satellite imagery. The proposed method, ExpDWT-VAE, introduces dual branches: one processes spatial domain input through convolutional operations, while the other extracts and processes frequency-domain features via 2D Haar wavelet decomposition, convolutional operation, and inverse DWT reconstruction. These branches merge to create an integrated spatial-frequency representation, further refined through convolutional and diagonal Gaussian mapping into a robust latent representation. We utilize a new satellite imagery dataset housed by the TerraFly mapping system to validate our method. Experimental results across several performance metrics highlight the efficacy of the proposed method at enhancing latent space representation.