Real time motion estimation using a neural architecture implemented on GPUs

This work describes a neural network based architecture that represents and estimates object motion in videos. This architecture addresses multiple computer vision tasks such as image segmentation, object representation or characterization, motion analysis and tracking. The use of a neural network architecture allows for the simultaneous estimation of global and local motion and the representation of deformable objects. This architecture also avoids the problem of finding corresponding features while tracking moving objects. Due to the parallel nature of neural networks, the architecture has been implemented on GPUs that allows the system to meet a set of requirements such as: time constraints management, robustness, high processing speed and re-configurability. Experiments are presented that demonstrate the validity of our architecture to solve problems of mobile agents tracking and motion analysis.     

Circulating antibodies against p53 protein in patients with ovarian carcinoma. Correlation with clinicopathologic features and survival

BACKGROUND: Genetic alterations of the p53 tumor suppressor protein are the most frequent molecular events in human carcinogenesis. For as yet unknown reasons, mutant p53 often acts as an immunogen for autoantibody generation. These autoantibodies can be detected in the serum of cancer patients. The presence of such antibodies has been identified in a subset of patients with ovarian carcinoma, but their clinical significance has not been investigated. METHODS: Serum samples from patients with ovarian carcinoma were quantitatively analyzed for the presence of p53 autoantibodies with a time-resolved immunofluorometric procedure. Tumor p53 overexpression was assessed by immunohistochemical analysis of tissue sections. Kaplan-Meier survival curves were calculated for p53 antibody positive and negative patients, and the Cox model was used to evaluate the strength of the associations between the presence of serum p53 antibodies and cancer relapse or death, and also between the presence of such antibodies and other clinicopathologic features. RESULTS: p53 antibodies were detected in the serum of 41 of 174 patients with ovarian carcinoma (24%). Antibody levels ranged from a few hundred to 9 x 10(6) arbitrary Units/L, and fluctuated during the course of the disease. p53 antibody positive patients tended to have tumors overexpressing p53, but the association between the two parameters was not statistically significant (P = 0.13). There was also no association between the presence of p53 antibodies and clinical stage, tumor histologic type, or overall patient survival. However, these antibodies were more frequently present in patients older than 50 years (P = 0.001), in patients with moderately or poorly differentiated tumors (P = 0.001), and in patients who received chemotherapy (P = 0.015), and who suffered relapse after surgery (P = 0.018). In univariate analysis, p53 antibody positive patients were at an increased risk for relapse but not death. In multivariate analysis, the differences in disease free and overall survival between patients who were p53 antibody positive or negative were not statistically significant. CONCLUSIONS: p53 autoantibodies are found frequently in the serum of patients with ovarian carcinoma. The presence of such autoantibodies was associated with older patient age, more aggressive tumors, and reduced patient disease free survival. In multivariate analysis the prognostic value of p53 autoantibodies was not statistically significant.     

Fast 2D/3D object representation with growing neural gas

This work presents the design of a real-time system to model visual objects with the use of self-organising networks. The architecture of the system addresses multiple computer vision tasks such as image segmentation, optimal parameter estimation and object representation. We first develop a framework for building non-rigid shapes using the growth mechanism of the self-organising maps, and then we define an optimal number of nodes without overfitting or underfitting the network based on the knowledge obtained from information-theoretic considerations. We present experimental results for hands and faces, and we quantitatively evaluate the matching capabilities of the proposed method with the topographic product. The proposed method is easily extensible to 3D objects, as it offers similar features for efficient mesh reconstruction.

     

Handling multiple materials for exposure of digital forgeries using 2-D lighting environments

The distribution of incident light is an important physics-based cue for exposing image manipulations. If an image has been composed from multiple sources, it is likely that the illumination environments of the spliced objects differ. Johnson and Farid introduced a proof-of-principle algorithm for a forensic comparison of lighting environments. However, this baseline approach suffers from relatively strict assumptions that limit its practical applicability. In this work, we address one of the biggest limitations, namely the need to compute a lighting environment from patches of homogeneous material. To compute a lighting environment from multiple-color surfaces, we propose a method that we call “intrinsic contour estimation” (ICE). ICE is able to integrate reflectances from multiple materials into one lighting environment, as long as surfaces of different materials share at least two similar normal vectors. We validate the proposed method in a controlled ground-truth experiment on two datasets, with light from three different directions. These experiments show that using ICE can improve the median estimation error by almost 50 %, and the mean error by almost 30 %.     

Prognostic implications of proliferating cell nuclear antigen (PCNA), AgNORs and P53 in non-Hodgkin's lymphomas

We investigated the prognostic value of proliferating cell nuclear antigen (PCNA) and p53 oncoprotein expression and of nucleolar organiser region (NOR) scoring, in relation to classic clinicopathological parameters, in a series of non-Hodgkin's lymphomas (NHL). Paraffin embedded tissue from 91 patients with NHL was stained immunohistochemically with the monoclonal antibodies PC-10 (PCNA) and DO-1 (p53) and histochemically with the AgNOR technique. The median follow-up was 48 (4 to 193) months. The impact of PCNA and p53 expression and of AgNOR number on survival was tested using univariate as well as multivariate analysis, in order to circumvent the heterogeneity in histologic grade, type and therapy. Univariate analysis identified seven variables related to overall survival: histologic type and grade, clinical stage, chemotherapy, p53 labelling index (LI), PCNA LI and AgNOR score, whereas only one parameter i.e. histologic grade influenced disease-free survival. In multivariate analysis stage, PCNA LI and AgNOR score predicted independently overall survival. PCNA was also the only independent predictor of post-relapse survival and histologic grade the most important indicator of disease-free survival. In conclusion, PCNA expression and AgNOR number may be better predictors of overall and post-relapse survival than histologic grade. The latter remains the most valuable prognostic indicator of disease-free survival.     

Hybrid deep convolutional neural models for iris image recognition

Multimedia Tools and Applications - This paper briefly explains about the application of deep learning-based methods for biometric applications. This work attempts to solve the problem of limited...     

Active Foreground Region Extraction and Tracking for Sports Video Annotation

Automatic video segmentation plays a vital role in sports videos annotation. This paper presents a fully automatic and computationally efficient algorithm for analysis of sports videos. Various methods of automatic shot boundary detection have been proposed to perform automatic video segmentation. These investigations mainly concentrate on detecting fades and dissolves for fast processing of the entire video scene without providing any additional feedback on object relativity within the shots. The goal of the proposed method is to identify regions that perform certain activities in a scene. The model uses some low-level feature video processing algorithms to extract the shot boundaries from a video scene and to identify dominant colours within these boundaries. An object classification method is used for clustering the seed distributions of the dominant colours to homogeneous regions. Using a simple tracking method a classification of these regions to active or static is performed. The efficiency of the proposed framework is demonstrated over a standard video benchmark with numerous types of sport events and the experimental results show that our algorithm can be used with high accuracy for automatic annotation of active regions for sport videos.