The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Social media is rocking the world in recent year, which makes modeling social media contents important. However, the heterogeneity of social media data is the main constraint. This paper focuses on inferring emotions from large-scale social media data. Tweets on social media platform, always containing heterogeneous information from different combinations of modalities, are utilized to construct a...
In this work, a quality estimation based multi-focus image fusion method (QEBIF) is proposed. In this method, the all-in-focus image is generated by pixel-wise summarizing the multi-focus images with their estimated focus levels as weights. Since the visual quality of an image is highly correlated with its focus level, the visual quality is estimated to be the pre-measurements of focus levels. Via...
Accurate pedestrian detection with high speed is always of great interests especially for practical application. Detectors usually follow the feature selection paradigm, and need to first construct rich and diverse features. In particular, current state-of-the-arts generate more channels of feature by convolving the basic feature channels with filter banks, which significantly improves accuracy. In...
We address the problem of automatically recognizing artistic movement in digitized paintings. We make the following contributions: Firstly, we introduce a large digitized painting database that contains refined annotations of artistic movement. Secondly, we propose a new system for the automatic categorization that resorts to image descriptions by color structure and novel topographical features as...
This paper presents a novel person re-identification framework based on data fusion. The pipeline of the proposed method is composed of two stages. First, a metric learning paradigm is applied on a bunch of distinct feature extractors to produce an ensemble of estimated distance measures, which are subsequently penalized according to their confidence in evidencing the correct matches from the false...
Learning visual attributes is an effective approach for zero-shot recognition. However, existing methods are restricted to learning explicitly nameable attributes and cannot tell which attributes are more important to the recognition task. In this paper, we propose a unified framework named Grouped Simile Ensemble (GSE). We claim our contributions as follows. 1) We propose to substitute explicit attribute...
We introduce T-LESS, a new public dataset for estimating the 6D pose, i.e. translation and rotation, of texture-less rigid objects. The dataset features thirty industry-relevant objects with no significant texture and no discriminative color or reflectance properties. The objects exhibit symmetries and mutual similarities in shape and/or size. Compared to other datasets, a unique property is that...
The ongoing development of Autonomous Ground Vehicle technologies necessitates for classification of terrain as road and off road to identify the drivable path and optimal velocity for traversal of the vehicle. Terrain consists of different texture types, classification of terrain into different classes is a difficult and challenging task. In this paper the feature set extraction and classification...
Presentation attacks such as printed iris images or patterned contact lenses can be used to circumvent an iris recognition system. Different solutions have been proposed to counteract this vulnerability with Presentation Attack Detection (commonly called liveness detection) being used to detect the presence of an attack, yet independent evaluations and comparisons are rare. To fill this gap we have...
Pill identification is a serious concern for pharmacists due to similarity of pill appearances. Pill imprints usually contain important information that can be used to add or search for pill information on existing pill databases. However, current techniques for extracting imprints often give results as vectors which cannot be used with existing databases. Thus, this paper proposed an approach for...
Plants are to be considered as one of the important things that plays a very essential role for all living beings exists on earth. But due to some unawareness and environment deterioration, some very rare plants are on the verge of extinction. Knowledge of rare leaves used for medicine and other plants is very critical in future. Leaf identification and classification plays a vital role for plant...
Assessment of aging civil infrastructure should be done periodically to getting information about the structural condition. In context to it, classification, detection, and localization of cracks within these concrete structures is of paramount importance. The most commonly used procedure, i.e. visual inspection, is executed manually by human inspectors, and thus, its accuracy depends on personnel's...
A large number of face recognition algorithms have been developed in last decades. Over the past four decades, performance of Face Recognition on frontal faces in controlled environment has improved significantly but frontal faces with uncontrolled environment and expression remains a challenge. The GBU Based Face Recognition Techniques focuses on the attention of the fundamental problem of comparing...
In this paper we present a novel preliminary study based on state of the art computation techniques applied on WCE video images for extracting features and patterns from different views of the same polyp. The extracted pattern and features are synthesized by using LG registration technique to create better views for each polyp by removing artifacts and noise. The outcome images (views) from each polyp...
Melanoma is the deadliest form of skin cancer, and its depth of invasion (DoI) is an important factor used by pathologist for grading the severity of skin disease. In this paper, we propose an automated technique for measuring melanoma DoI in MART1 stained skin histopathological images. The proposed technique first segments skin melanoma areas based on image color features. The skin epidermis is then...
Detection of nuclei is an important step in phenotypic profiling of histology sections that are usually imaged in bright field. However, nuclei can have multiple phenotypes, which are difficult to model. It is shown that convolutional neural networks (CNN)s can learn different phenotypic signatures for nuclear detection, and that the performance is improved with the feature-based representation of...
Abstract—In this paper, we analyse the use of Convolutional Neural Networks (CNNs or ConvNets) to discriminate vegetation species with few labelled samples. To the best of our knowledge, this is the first work dedicated to the investigation of the use of deep features in such task. The experimental evaluation demonstrate that deep features significantly outperform wellknown feature extraction techniques...
This paper presents a novel method for theautomation of highlight extraction using broadcastcricket video. The top-down hierarchical approachyielded an average frame processing speed of 0.04seconds. The Motion History Image (MHI) method wasused to detect the camera zoom-in motion which is asemantic feature of the bowler run-up sequence. A multispatialapproach to feature extraction maximizedhighlight...
Sparse Representation-based Classifier (SRC) is less sensitive to the shortage of data and the selection of feature space. In this paper, SRC is adopted to perform automatic analysis of tongue substance color and coating color which is considered as small dataset classification task. Firstly, for both training samples and testing samples, the tongue body regions are segmented, the regions of tongue...
This paper proposed a framework based on reinforcementlearning for color object segmentation in videosequences. This agent-based method is used to find the appropriatevalues to detect an object. The reinforcement learningagent contains some sample images and their ground truthfrom which to learn. The information obtained from theexploration/exploitation of the solution space is used by theintelligent...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.