WINE: Wavelet-Guided GAN Inversion and Editing for High-Fidelity Refinement
This addresses a specific bottleneck in GAN inversion for image editing and refinement, offering incremental improvements in high-fidelity reconstruction.
The paper tackles the challenge of accurately reconstructing image-specific details in GAN inversion, which often biases toward low-frequency information, by proposing WINE, a wavelet-guided model that transfers high-frequency information through wavelet coefficients. The results show WINE preserves high-frequency details and enhances image quality, outperforming state-of-the-art models in editing scenarios with a balance between editability and reconstruction quality.
Recent advanced GAN inversion models aim to convey high-fidelity information from original images to generators through methods using generator tuning or high-dimensional feature learning. Despite these efforts, accurately reconstructing image-specific details remains as a challenge due to the inherent limitations both in terms of training and structural aspects, leading to a bias towards low-frequency information. In this paper, we look into the widely used pixel loss in GAN inversion, revealing its predominant focus on the reconstruction of low-frequency features. We then propose WINE, a Wavelet-guided GAN Inversion aNd Editing model, which transfers the high-frequency information through wavelet coefficients via newly proposed wavelet loss and wavelet fusion scheme. Notably, WINE is the first attempt to interpret GAN inversion in the frequency domain. Our experimental results showcase the precision of WINE in preserving high-frequency details and enhancing image quality. Even in editing scenarios, WINE outperforms existing state-of-the-art GAN inversion models with a fine balance between editability and reconstruction quality.