There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

One of the main challenges in feature learning using Deep Convolutional Neural Networks (DCNNs) for large-scale face recognition is the design of appropriate loss functions that can enhance the discriminative power. Centre loss penalises the distance between deep features and their corresponding class centres in the Euclidean space to achieve intra-class compactness. SphereFace assumes that the linear transformation matrix in the last fully connected layer can be used as a representation of the class centres in the angular space and therefore penalises the angles between deep features and their corresponding weights in a multiplicative way. Recently, a popular line of research is to incorporate margins in well-established loss functions in order to maximise face class separability. In this paper, we propose an Additive Angular Margin Loss (ArcFace) to obtain highly discriminative features for face recognition. The proposed ArcFace has a clear geometric interpretation due to its exact correspondence to geodesic distance on a hypersphere. We present arguably the most extensive experimental evaluation against all recent state-of-the-art face recognition methods on ten face recognition benchmarks which includes a new large-scale image database with trillions of pairs and a large-scale video dataset. We show that ArcFace consistently outperforms the state of the art and can be easily implemented with negligible computational overhead. To facilitate future research, the code has been made available.

/pdf/arcface-additive-angular-margin-loss-for-deep-face-7vvlem659h.pdf

ArcFace: Additive Angular Margin Loss for Deep Face Recognition

The PASCAL Visual Object Classes Challenge

Automated analysis of human affective behavior has attracted increasing attention from researchers in psychology, computer science, linguistics, neuroscience, and related disciplines. However, the existing methods typically handle only deliberately displayed and exaggerated expressions of prototypical emotions despite the fact that deliberate behaviour differs in visual appearance, audio profile, and timing from spontaneously occurring behaviour. To address this problem, efforts to develop algorithms that can process naturally occurring human affective behaviour have recently emerged. Moreover, an increasing number of efforts are reported toward multimodal fusion for human affect analysis including audiovisual fusion, linguistic and paralinguistic fusion, and multi-cue visual fusion based on facial expressions, head movements, and body gestures. This paper introduces and surveys these recent advances. We first discuss human emotion perception from a psychological perspective. Next we examine available approaches to solving the problem of machine understanding of human affective behavior, and discuss important issues like the collection and availability of training and test data. We finally outline some of the scientific and engineering challenges to advancing human affect sensing technology.

A Survey of Affect Recognition Methods: Audio, Visual, and Spontaneous Expressions

In this paper, we exploit the advantages of tensor representations and propose a Supervised Multilinear Learning Model for regression. The model is based on the Canonical (CAN-DECOMP)/Parallel Factors (PARAFAC) decomposition of tensors of multiple modes and allows the simultaneous projection of an input tensor to more than one discriminative directions along each mode. These projection weights are obtained by optimizing a ϵ-insensitive loss functions which leads to generalized Support Tensor Regression (STR). The methods are validated on the problems of head pose estimation using real data from publicly available databases.

/pdf/higher-order-support-tensor-regression-for-head-pose-d4g3fq8pdv.pdf

Higher order support tensor regression for head pose estimation

In this paper we present a novel similarity measure between two image sequences, that is a)robust to different viewpoints and recording conditions (illumination variations and clothing) b)robust to geometric transformations (translation, scale and rotation transformations) and c)invariant to the number of frames of the image sequence as well as of its time scaling. More precisely, we create a similarity measure that exploits the underlying relationships among neighborhoods of detected spatiotemporal points in a frame of an image sequence. We find the space in which the similarities of neighboring spatiotemporal points lie in, and map it to another space of smaller dimensionality. In the new space the projected similarities are of fixed dimensionality, depending on the number of neighbors we have considered. We use the information about that newly extracted space to define a novel similarity measure between two image sequences and create in that way a similarity vector that can be used as an input to a classifier. We apply the proposed similarity measure to the 'action pair matching' problem, in which we try to decide whether two action image sequences contain the same action or not. Experiments conducted using the Action Similarity Labeling (ASLAN) dataset verify the superiority of the proposed method over state of the art techniques in terms of accuracy rate.

Exploring the similarities of neighboring spatiotemporal points for action pair matching

Nowadays, both computer vision researchers and psychology experts show an increased interest for human facial expression analysis. Despite the huge amount of research that has been dedicated to this area, almost all of them concern data recorded in controlled laboratory conditions, which does not always reflect the real world environment in which the human face is partially occluded. Six basic facial expressions are investigated in that case, i.e.when the eyes and eyebrows or the mouth regions are left out. We are interested in finding which part of the face comprised sufficient information with respect to the entire face, in order to correctly classify these six expressions. Each image from the two databases used is convolved with a set of Gabor filters having various orientations and frequencies. The new feature vectors are classified using a maximum correlation classifier and the cosine similarity measure approaches. Overall, the method provides robustness against partial occlusion.

Recognition of facial expressions in presence of partial occlusion

A novel method based on shape and texture information is proposed in this paper for facial expression recognition from video sequences. The Discriminant Non-negative Matrix Factorization (DNMF) algorithm is applied at the image cor-responding to the greatest intensity of the facial expression (last frame of the video sequence), extracting that way the texture information. A Support Vector Machines (SVMs) system is used for the classification of the shape information derived from tracking the Candide grid over the video sequence. The shape information consists of the differences of the node coordinates between the first (neutral) and last (fully expressed facial expression) video frame. Subsequently, fusion of texture and shape information obtained is performed using Radial Basis Function (RBF) Neural Net-works (NNs). The accuracy achieved is equal to 98,2% when recognizing the six basic facial expressions.

/pdf/facial-expression-recognition-using-shape-and-texture-3obzrnbwp7.pdf

Facial expression recognition using shape and texture information

In this paper we address the action spotting problem, that is the spatiotemporal detection and localization of an action. We first calculate a novel objective function between an input video sequence and an action's weights tensor, as acquired from a Support Tensor Machine classifier. We subsequently search for an appropriate transformation that maximizes the objective function, calculated as the multiplication of the original input tensor with the weights tensor. The proposed algorithm is very fast, as the above mentioned multiplication involves a set of a separable filters applied along each mode. We demonstrate the effectiveness of our method with experiments in a publicly available database where we show that our method outperforms existing techniques in terms of spatiotemporal action localization.

Irene Kotsia

Papers

Higher order support tensor regression for head pose estimation

Exploring the similarities of neighboring spatiotemporal points for action pair matching

Recognition of facial expressions in presence of partial occlusion

Facial expression recognition using shape and texture information

Support tensor action spotting