Look globally, age locally: Face aging with an attention mechanism
This work addresses face aging for cross-age recognition and entertainment applications, representing an incremental improvement over existing cGANs-based methods.
The paper tackled the problem of ghosting and blurriness in face aging using conditional GANs by introducing an attention mechanism to focus only on aging-relevant regions, resulting in superior performance on the Morph dataset in image quality, identity preservation, and age accuracy.
Face aging is of great importance for cross-age recognition and entertainment-related applications. Recently, conditional generative adversarial networks (cGANs) have achieved impressive results for face aging. Existing cGANs-based methods usually require a pixel-wise loss to keep the identity and background consistent. However, minimizing the pixel-wise loss between the input and synthesized images likely resulting in a ghosted or blurry face. To address this deficiency, this paper introduces an Attention Conditional GANs (AcGANs) approach for face aging, which utilizes attention mechanism to only alert the regions relevant to face aging. In doing so, the synthesized face can well preserve the background information and personal identity without using the pixel-wise loss, and the ghost artifacts and blurriness can be significantly reduced. Based on the benchmarked dataset Morph, both qualitative and quantitative experiment results demonstrate superior performance over existing algorithms in terms of image quality, personal identity, and age accuracy.