Search

In this work, we adhere to explore a Multi-Tasking learning (MTL) based network to perform document attribute classification such as the font type, font size, font emphasis and scanning resolution classification of a document image. To accomplish these tasks, we operate on either segmented word level or on uniformed size patches randomly cropped out of the document. Furthermore, a hybrid convolution neural network (CNN) architecture "MTL+MI", which is based on the combination of MTL and Multi-Instance (MI) of patch and word is used to accomplish joint learning for the classification of the same document attributes. The contribution of this paper are three fold: firstly, based on segmented word images and patches, we present a MTL based network for the classification of a full document image. Secondly, we propose a MTL and MI (using segmented words and patches) based combined CNN architecture ("MTL+MI") for the classification of same document attributes. Thirdly, based on the multi-tasking classifications of the words and/or patches, we propose an intelligent voting system which is based on the posterior probabilities of each words and/or patches to perform the classification of document's attributes of complete document image., Comment: 12 pages, 6 figures, 10 tables

The method of using Hopf algebras for calculating Feynman integrals developed by Abreu et al. is applied to the two-loop non-planar on-shell diagram with massless propagators and three external mass scales. We show that the existence of the method of cut Feynman diagrams comprising of the coproduct, the first entry condition and integrability condition that was found to be true for the planar case also holds for the non-planar case; furthermore, the non-planar symbol alphabet is the same as for the planar case. This is one of the main results of this work, and they have been obtained by a systematic analysis of the relevant cuts, using the symbolic manipulation codes HypExp and PolyLogTools. The obtained result for the symbol is cross-checked by an analysis of the known two-loop original Feynman integral result. In addition, we also reconstruct the full result from the symbol. This is the other main result in this paper., Comment: 26 pages, 5 main figures, 1 table, 1 footnote; Compared to v1, many improvements in discussion. A new section (sec. 8) has been added which now gives the full result. Discussion and conclusions extended and improved. Abstract and introduction also amended to account for the new calculation. New references added

Multi-threaded applications are capable of exploiting the full potential of many-core systems. However, Network-on-Chip (NoC) based inter-core communication in many-core systems is responsible for 60-75% of the miss latency experienced by multi-threaded applications. Delay in the arrival of critical data at the requesting core severely hampers performance. This brief presents some interesting insights about how critical data is requested from the memory by multi-threaded applications. Then it investigates the cause of delay in NoC and how it affects the performance. Finally, this brief shows how NoC-aware memory access optimisations can significantly improve performance. Our experimental evaluation considers early restart memory access optimisation and demonstrates that by exploiting NoC resources, critical data can be prioritised to reduce miss penalty by 10-12% and improve system performance by 7-11%.

Over the last decades, the world has been witnessing growing threats to the security in urban spaces, which has augmented the relevance given to visual surveillance solutions able to detect, track and identify persons of interest in crowds. In particular, unmanned aerial vehicles (UAVs) are a potential tool for this kind of analysis, as they provide a cheap way for data collection, cover large and difficult-to-reach areas, while reducing human staff demands. In this context, all the available datasets are exclusively suitable for the pedestrian re-identification problem, in which the multi-camera views per ID are taken on a single day, and allows the use of clothing appearance features for identification purposes. Accordingly, the main contributions of this paper are two-fold: 1) we announce the UAV-based P-DESTRE dataset, which is the first of its kind to provide consistent ID annotations across multiple days, making it suitable for the extremely challenging problem of person search, i.e., where no clothing information can be reliably used. Apart this feature, the P-DESTRE annotations enable the research on UAV-based pedestrian detection, tracking, re-identification and soft biometric solutions; and 2) we compare the results attained by state-of-the-art pedestrian detection, tracking, reidentification and search techniques in well-known surveillance datasets, to the effectiveness obtained by the same techniques in the P-DESTRE data. Such comparison enables to identify the most problematic data degradation factors of UAV-based data for each task, and can be used as baselines for subsequent advances in this kind of technology. The dataset and the full details of the empirical evaluation carried out are freely available at http://p-destre.di.ubi.pt/., Comment: 11 pages, 12 figures, 7 tables

The ASPIRE program, which is based on the Landau singularities and the method of power geometry to unveil the regions required for the evaluation of a given Feynman diagram asymptotically in a given limit, also allows for the evaluation of scaling coming from the top facets. In this work, we relate the scaling having equal components of the top facets of the Newton polytope to the maximal cut of given Feynman integrals. We have therefore connected two independent approaches to the analysis of Feynman diagrams., Comment: 34 pages latex, 7 figures and 2 tables, ancillary Mathematica files provided, substantially extended and improved, added new sections(sub-section-2.5, section-3,4,appendix-B), References added, version accepted for publication in EPJC