Search for: All records

Video Paragraph Captioning aims to generate a multi-sentence description of an untrimmed video with multiple temporal event locations in a coherent storytelling. Following the human perception process, where the scene is effectively understood by decomposing it into visual (e.g. human, animal) and non-visual components (e.g. action, relations) under the mutual influence of vision and language, we first propose a visual-linguistic (VL) feature. In the proposed VL feature, the scene is modeled by three modalities including (i) a global visual environment; (ii) local visual main agents; (iii) linguistic scene elements. We then introduce an autoregressive Transformer-in-Transformer (TinT) to simultaneously capture the semantic coherence of intra- and inter-event contents within a video. Finally, we present a new VL contrastive loss function to guarantee the learnt embedding features are consistent with the captions semantics. Comprehensive experiments and extensive ablation studies on the ActivityNet Captions and YouCookII datasets show that the proposed Visual-Linguistic Transformer-in-Transform (VLTinT) outperforms previous state-of-the-art methods in terms of accuracy and diversity. The source code is made publicly available at: https://github.com/UARK-AICV/VLTinT. 

The automatic classification of electrocardiogram (ECG) signals has played an important role in cardiovascular diseases diagnosis and prediction. Deep neural networks (DNNs), particularly Convolutional Neural Networks (CNNs), have excelled in a variety of intelligent tasks including biomedical and health informatics. Most the existing approaches either partition the ECG time series into a set of segments and apply 1D-CNNs or divide the ECG signal into a set of spectrogram images and apply 2D-CNNs. These studies, however, suffer from the limitation that temporal dependencies between 1D segments or 2D spectrograms are not considered during network construction. Furthermore, meta-data including gender and age has not been well studied in these researches. To address those limitations, we propose a multi-module Recurrent Convolutional Neural Networks (RCNNs) consisting of both CNNs to learn spatial representation and Recurrent Neural Networks (RNNs) to model the temporal relationship. Our multi-module RCNNs architecture is designed as an end-to-end deep framework with four modules: (i) timeseries module by 1D RCNNs which extracts spatio-temporal information of ECG time series; (ii) spectrogram module by 2D RCNNs which learns visual-temporal representation of ECG spectrogram ; (iii) metadata module which vectorizes age and gender information; (iv) fusion module which semantically fuses the information from three above modules by a transformer encoder. Ten-fold cross validation was used to evaluate the approach on the MIT-BIH arrhythmia database (MIT-BIH) under different network configurations. The experimental results have proved that our proposed multi-module RCNNs with transformer encoder achieves the state-of-the-art with 99.14% F1 score and 98.29% accuracy. 

This paper is on an Eulerian-Eulerian (EE) approach that utilizes Godunov’s scheme to deal with a running shock that interacts with a cloud of particles. The EE approach treats both carrier phase (fluid phase) and dispersed phase (particle phase) in the Eulerian frame. In this work, the fluid equations are the Euler equations for the compressible gas while the particle equations are based on a recently developed model to solve for the number density, velocity, temperature, particle sub-grid scale stresses, and particle sub-grid scale heat fluxes. The carrier and dispersed phases exchange momentum and heat, which are modeled through incorporating source terms in their equations. Carrier and dispersed phase equation form a hyperbolic set of differential equations, which are numerically solved with Godunov’s scheme. The numerical solutions are obtained in this work for a two-dimensional normal running shock interacting with a rectangular cloud of particles. The results generated by the EE approach were compared against the results that were generated by a well-stablished Eulerian-Lagragian (EL) approach that treats the carrier phase in an Eulerian frame, while does the dispersed phase in a Lagrangian framework where individuals particles are traced and solved. For the considered configuration, the EE approach reproduced the EL results with a very good accuracy. 

Abstract Photosynthetic traits suggest that shade tolerance may explain the contrasting success of two conifer taxa, Podocarpaceae and Pinaceae, in tropical forests. Needle‐leaved species fromPinus(Pinaceae) are generally absent from tropical forests, whereasPinus krempfii, a flat‐leaved pine, and numerous flat‐leaved Podocarpaceae are abundant. Respiration (R) traits may provide additional insight into the drivers of the contrasting success of needle‐ and flat‐leaved conifers in tropical forests.We measured the short‐term respiratory temperature (RT) response between 10 and 50°C and foliar morphological traits of three needle‐ and seven flat‐leaved conifer species coexisting in a tropical montane forest in the Central Highlands of Vietnam containing notable conifer diversity. We fit a lognormal polynomial model to each RT curve and extracted the following three parameters:a(basalR), andbandc(together describing the shape of the response).Needle‐leaved species (Pinus kesiya,Pinus dalatensisandDacrydium elatum) had higher rates of area‐basedRat 25°C (R_25‐area) as well as higher area‐based modelled basal respiration (a) than flat‐leaved species (P. krempfii,Podocarpus neriifolius,Dacrycarpus imbricatus,Nageia nana,Taxus wallichiana,Keteeleria evelynianaandFokienia hodginsii). No significant differences were found between needle‐ and flat‐leaved species in mass‐basedR₂₅(R_25‐mass) or in the shape of the RT response (bandc); however, interspecific differences inR_25‐mass,Rat nighttime temperature extremes (R_4.1andR_20.6) and leaf traits were apparent.Differences inR_25‐areaandasuggest that needle‐leaved foliage may be more energetically costly to maintain than flat‐leaved foliage, providing new insight and additional support for the hypothesis that shade tolerance is an important driver of Podocarpaceae success and Pinaceae absence in the majority of tropical forests.Interspecific differences inR_25‐massand leaf traits highlight that varying ecological strategies are employed by conifers to coexist and survive in the Central Highlands of Vietnam. Ultimately, these data further our understanding of current conifer biogeographical distributions and underscore the need for additional studies to elucidate the effects of extreme temperature events on the continued survival of conifers in this unique forest. A freePlain Language Summarycan be found within the Supporting Information of this article.