Deep fisher faces

Abstract

Most current state-of-the-art methods for unconstrained face recognition use deep convolutional neural networks. Recently, it has been proposed to augment the typically used softmax cross-entropy loss by adding a center loss trying to minimize the distance between the face images and their class centers. In this work we further extend the center (intra-class) loss with an inter-class loss reminiscent of the popular early face recognition approach Fisherfaces. To this end we add a term that directly optimizes the distances of the class centers appearing in a batch in dependence of the input images. We evaluate the new loss on two popular databases for unconstrained face recognition, the Labeled Faces in the Wild and the Youtube Faces database. In both cases the new loss achieves competitive results.

Multi-attentional Deepfake Detection

“…To achieve these goals, we propose a Region Independence Loss which helps to reduce the overlap among attention maps and keep the consistency for different inputs. We apply BAP on the pooled feature map D otained in Section 3.2 to get a "semantic feature vector" : V ∈ R M ×N , and the Regional Independence Loss is defined as below by modifying the center loss in [15]:…”

ELoPE: Fine-Grained Visual Classification with Efficient Localization, Pooling and Embedding

“…CNNs are therefore often used to train a discriminative embedding space in which face images can be compared efficiently and accurately. The embeddings are learned using specifically designed loss functions such as center loss [30], triplet loss [23] or DFF [10]. We insert such an embedding layer trained with a loss function based similar to [10] into the backbone CNN as penultimate layer.…”

“…The embeddings are learned using specifically designed loss functions such as center loss [30], triplet loss [23] or DFF [10]. We insert such an embedding layer trained with a loss function based similar to [10] into the backbone CNN as penultimate layer. We show that this greatly improves the performance of the softmax classifier.…”

Deep multi-factor forensic face recognition

“…There are many other alternative methods, including the range loss in [117], fisher face in [118], marginal loss in [120], sphere face in [121], etc. Each of these methods has its own uniqueness and advantages under certain setup.…”

“…The first way is to enhance the generalization and discriminative capability of representation model. Examples include range loss [117], fisher face [118], center invariant loss [119], marginal loss [120], sphere face [121], etc. The second way is to improve the estimation of partitions in the feature space.…”