Age-Oriented Face Synthesis With Conditional Discriminator Pool and Adversarial Triplet Loss

Abstract

The vanilla Generative Adversarial Networks (GAN) are commonly used to generate realistic images depicting aged and rejuvenated faces. However, the performance of such vanilla GANs in the age-oriented face synthesis task is often compromised by the mode collapse issue, which may result in the generation of faces with minimal variations and a poor synthesis accuracy. In addition, recent age-oriented face synthesis methods use the L1 or L2 constraint to preserve the identity information on synthesized faces, which implicitly limits the identity permanence capabilities when these constraints are associated with a trivial weighting factor. In this paper, we propose a method for the age-oriented face synthesis task that achieves a high synthesis accuracy with strong identity permanence capabilities. Specifically, to achieve a high synthesis accuracy, our method tackles the mode collapse issue with a novel Conditional Discriminator Pool (CDP), which consists of multiple discriminators, each targeting one particular age category. To achieve strong identity permanence capabilities, our method uses a novel Adversarial Triplet loss. This loss, which is based on the Triplet loss, adds a ranking operation to further pull the positive embedding towards the anchor embedding resulting in significantly reduced intra-class variances in the feature space. Through extensive experiments, we show that our proposed method outperforms state-of-the-art methods in terms of synthesis accuracy and identity permanence capabilities, qualitatively and quantitatively.

Hybrid Spatial-Channel Attention Mechanism for Cross-Age Face Recognition

“…where the identity feature decomposed by the model is X id and the real identity label is Y id , Y id i is the real identity label corresponding to the i-th sample, and X id i is the identity information obtained from the model decomposition [46]. From the above, we define the total loss function as L total , and the overall loss function can be derived as follows:…”

Deep Clustering via Center-Oriented Margin Free-Triplet Loss for Skin Lesion Detection in Highly Imbalanced Datasets

“…To maintain discriminability, images belonging to the same class should be located in close proximity compared to images of opposing classes. A common approach for learning discriminative embeddings is based on the triplet loss that aims to maximize across-cluster over withincluster distances [23], [50]. Calculation of triplet loss involves selection of multiple instances of image triplets, where each instance contains an anchor image (A) along with a positive image (P) from the same class, and a negative image (N) from the opposite class.…”