Diana Borza

Cloud SLA negotiation for energy saving — A particle swarm optimization approach

The necessity of balancing the obtained performance with the energy consumed is an emerging ambition for cloud computing research. Performance in cloud computing is defined through Service Level Agreement contracts between the cloud provider and cloud customer, being a projection of the customers perspective on the service offered by the cloud provider. Although more and more research efforts go into standardizing Service Level Agreement in cloud systems, the area is still at its early ages. This paper proposes a Service Level Agreement negotiation protocol based on particle swarm optimization techniques, for obtaining a balance between the energy consumed and performance offered in the cloud. The two parties of the defined negotiation protocol are the performance-oriented cloud customer and the energy-oriented cloud provider. The agreement resulted from the negotiation process satisfies the two major negotiation properties we aim for: closeness to Pareto optimality and high social welfare.

Eyeglasses lens contour extraction from facial images using an efficient shape description.

This paper presents a system that automatically extracts the position of the eyeglasses and the accurate shape and size of the frame lenses in facial images. The novelty brought by this paper consists in three key contributions. The first one is an original model for representing the shape of the eyeglasses lens, using Fourier descriptors. The second one is a method for generating the search space starting from a finite, relatively small number of representative lens shapes based on Fourier morphing. Finally, we propose an accurate lens contour extraction algorithm using a multi-stage Monte Carlo sampling technique. Multiple experiments demonstrate the effectiveness of our approach.

Automatic extrinsic camera parameters calibration using convolutional neural networks

Camera calibration is essential for accurate computer vision, and automatic calibration of some extrinsic parameters is needed in case the camera is placed on a mobile platform. The pitch and yaw angles, which are the most likely ones to change as the vehicle moves, can be inferred from the image coordinates of the vanishing point (VP). In this paper we present an artificial neural network approach for detecting the vanishing point position in road traffic scenarios. The network is trained using 2500 images which are first automatically annotated using a classical vanishing point detection algorithm, and then manually validated. The training and test datasets are made publicly available. The trained network was tested on more than 250 images not previously seen by the network, locating the VP accurately in more than 90% of the cases.

Real-Time Detection and Measurement of Eye Features from Color Images

The accurate extraction and measurement of eye features is crucial to a variety of domains, including human-computer interaction, biometry, and medical research. This paper presents a fast and accurate method for extracting multiple features around the eyes: the center of the pupil, the iris radius, and the external shape of the eye. These features are extracted using a multistage algorithm. On the first stage the pupil center is localized using a fast circular symmetry detector and the iris radius is computed using radial gradient projections, and on the second stage the external shape of the eye (of the eyelids) is determined through a Monte Carlo sampling framework based on both color and shape information. Extensive experiments performed on a different dataset demonstrate the effectiveness of our approach. In addition, this work provides eye annotation data for a publicly-available database.

A genetic-inspired negotiation algorithm for QoS and energy consumption tradeoffs in virtualized service centers

This paper proposes a genetic inspired algorithm for negotiating the tradeoffs between the workload Quality of Service requests and the service center computing resources energy consumption with the goal of allocating the service center computing resources in an energy efficient manner. The bilateral negotiation algorithm has two main parties: the workload tasks Quality of Service request, as a client, and the service center servers available computing resources, as a provider. Both the provider and the client are represented by agents and their offers/requests are modeled as chromosomes. A chromosome gene represents the value of the computing resources subject of negotiation. The genetic inspired negotiation process has an initial phase and a bargaining phase. In the initial phase, an initial chromosome population is generated for both the provider and the client and the values of their associated goal chromosomes are set. In the bargaining phase, the client and provider chromosomal populations are evolved using a cognitive process similar to the genetic evolution. An agreement is reached when the distance between one of the received offer/request chromosomes and a corresponding goal chromosome is below a predefined threshold.

Automatic Skin Tone Extraction for Visagism Applications.

In this paper we propose a skin tone classification system on three skin colors: dark, medium and light. We work on two methods which don’t require any camera or color calibration. The first computes color histograms in various color spaces on representative facial sliding patches that are further combined in a large feature vector. The dimensionality of this vector is reduced using Principal Component Analysis a Support Vector Machine determines the skin color of each region. The skin tone is extrapolated using a voting schema. The second method uses Convolutional Neural Networks to automatically extract chromatic features from augmented sets of facial images. Both algorithms were trained and tested on publicly available datasets. The SVM method achieves an accuracy of 86.67%, while the CNN approach obtains an accuracy of 91.29%. The proposed system is developed as an automatic analysis module in an optical visagism system where the skin tone is used in an eyewear virtual try-on software that allows users to virtually try glasses on their face using a mobile device with a camera. The system proposes only esthetically and functionally fit frames to the user, based on some facial features –skin tone included.

A Deep Learning Approach to Hair Segmentation and Color Extraction from Facial Images

In this paper we tackle the problem of hair analysis in unconstrained images. We propose a fully convolutional, multi-task neural network to segment the image pixels into hair, face and background classes. The network also decides if the person is bald or not. The detected hair pixels are analyzed by a color recognition module which uses color features extracted at super-pixel level and a Random Forest Classifier to determine the hair tone (black, blond, brown, red or white grey). To train and test the proposed solution, we manually segment more than 3500 images from a publicly available dataset. The proposed framework was evaluated on three public databases. The experiments we performed together with the hair color recognition rate of 92% demonstrate the efficiency of the proposed solution.

In the Eye of the Deceiver: Analyzing Eye Movements as a Cue to Deception

Deceit occurs in daily life and, even from an early age, children can successfully deceive their parents. Therefore, numerous book and psychological studies have been published to help people decipher the facial cues to deceit. In this study, we tackle the problem of deceit detection by analyzing eye movements: blinks, saccades and gaze direction. Recent psychological studies have shown that the non-visual saccadic eye movement rate is higher when people lie. We propose a fast and accurate framework for eye tracking and eye movement recognition and analysis. The proposed system tracks the position of the iris, as well as the eye corners (the outer shape of the eye). Next, in an offline analysis stage, the trajectory of these eye features is analyzed in order to recognize and measure various cues which can be used as an indicator of deception: the blink rate, the gaze direction and the saccadic eye movement rate. On the task of iris center localization, the method achieves within pupil localization in 91.47% of the cases. For blink localization, we obtained an accuracy of 99.3% on the difficult EyeBlink8 dataset. In addition, we proposed a novel metric, the normalized blink rate deviation to stop deceitful behavior based on blink rate. Using this metric and a simple decision stump, the deceitful answers from the Silesian Face database were recognized with an accuracy of 96.15%.

Real-time micro-expression detection from high speed cameras

This work presents an original real time, robust micro-expression detection algorithm. The algorithm analyses the movement modifications that occur around the most prominent facial regions using two absolute frame differences. Next, a machine learning algorithm is used to predict if a micro-expression occurred at a given frame t. Two classifiers were evaluated: decision tree and random forest classifier. The robustness of the proposed solution is increased by further processing the preliminary predictions of the classifier: the appropriate predicted micro-expression intervals are merged together and the interval that are too short are filtered out. The proposed solution achieved an 86.95% true positive rate on CASME2 dataset. The mean execution time of the proposed solution on 640×480 images is 9 milliseconds.

Automatic gender recognition for “in the wild” facial images using convolutional neural networks

Automatic recognition of human demographical attributes has implications in a variety of domains, such as surveillance systems, human computer interaction, marketing etc. In this paper, we present an automatic gender recognition method from facial images based on convolutional neural networks. In order to train the network, we merged together several face databases and also gathered and annotated a ∼70000 facial images from the internet. We trained, evaluated and compared several network architectures that achieved impressive results on other computer vision tasks. The best accuracy is obtained using Inception-v4 network: 98.2% on our dataset, and 84% on Adience dataset.