scholarly journals Exploring Convolutional Recurrent architectures for anomaly detection in videos: a comparative study

2021 ◽  
Vol 1 (1) ◽  
Author(s):  
Ambareesh Ravi ◽  
Fakhri Karray
Keyword(s):  
Anomaly Detection ◽  
Visual Learning ◽  
Trade Off ◽  
Computational Burden ◽  
Convolutional Networks ◽  
Learning Tasks ◽  
Visual Evidence ◽  
Spatio Temporal ◽  
Temporal Learning ◽  
Better Than

AbstractConvolutional Recurrent architectures are currently preferred for spatio-temporal learning tasks in videos to the 3D convolutional networks which accompany a huge computational burden and it is imperative to understand the working of different architectural configurations. But most of the current works on visual learning, especially for video anomaly detection, predominantly employ ConvLSTM networks and focus less on other possible variants of Convolutional Recurrent configurations for temporal learning which warrants a need to study the different possible variants to make informed, optimal design choices according to the nature of the application at hand. We explore a variety of Convolutional Recurrent architectures and the influence of hyper-parameters on their performance for the task of anomaly detection. Through this work, we also intend to quantify the efficiency of the architectures based on the trade-off between their performance and computational complexity. With comprehensive quantitative and visual evidence, we establish that the ConvGRU based configurations are the most effective and perform better than the popular ConvLSTM configurations on video anomaly detection tasks, in contrast to what is seen from the literature.

Making the most of individual differences in joint decisions

2018 ◽  
Author(s):  
Bahador Bahrami
Keyword(s):  
Decision Making ◽  
Direct Evidence ◽  
Cross Cultural ◽  
Collective Decision Making ◽  
Wisdom Of Crowds ◽  
Trade Off ◽  
Adaptive Value ◽  
Joint Decisions ◽  
Cultural Data ◽  
Better Than

Evidence for and against the idea that “two heads are better than one” is abundant. This chapter considers the contextual conditions and social norms that predict madness or wisdom of crowds to identify the adaptive value of collective decision-making beyond increased accuracy. Similarity of competence among members of a collective impacts collective accuracy, but interacting individuals often seem to operate under the assumption that they are equally competent even when direct evidence suggest the opposite and dyadic performance suffers. Cross-cultural data from Iran, China, and Denmark support this assumption of similarity (i.e., equality bias) as a sensible heuristic that works most of the time and simplifies social interaction. Crowds often trade off accuracy for other collective benefits such as diffusion of responsibility and reduction of regret. Consequently, two heads are sometimes better than one, but no-one holds the collective accountable, not even for the most disastrous of outcomes.

scholarly journals Spatio-Temporal Learning of Basketball Offensive Strategies

2015 ◽  
Author(s):  
Ching-Hang Chen ◽  
Tyng-Luh Liu ◽  
Yu-Shuen Wang ◽  
Hung-Kuo Chu ◽  
Nick C. Tang ◽  
...  
Keyword(s):  
Spatio Temporal ◽  
Temporal Learning ◽  
Offensive Strategies

scholarly journals The aerodynamic puzzle of nasalized fricatives: Aerodynamic and perceptual evidence from Scottish Gaelic

2015 ◽  
Vol 6 (2) ◽  
Author(s):  
Natasha Warner ◽  
Daniel Brenner ◽  
Jessamyn Schertz ◽  
Andrew Carnie ◽  
Muriel Fisher ◽  
...  
Keyword(s):  
Scottish Gaelic ◽  
Trade Off ◽  
Instrumental Data ◽  
The World ◽  
Perceptual Evidence ◽  
Better Than ◽  
Do So

AbstractScottish Gaelic is sometimes described as having nasalized fricatives (/ṽ/ distinctively, and [f̃, x̃, h̃], etc. through assimilation). However, there are claims that it is not aerodynamically possible to open the velum for nasalization while maintaining frication noise. We present aerodynamic data from 14 native Scottish Gaelic speakers to determine how the posited nasalized fricatives in this language are realized. Most tokens demonstrate loss of nasalization, but nasalization does occur in some contexts without aerodynamic conflict, e.g., nasalization with the consonant realized as an approximant, nasalization of [h̃], nasalization on the preceding vowel, or sequential frication and nasalization. Furthermore, a very few tokens do contain simultaneous nasalization and frication with a trade-off in airflow. We also present perceptual evidence showing that Gaelic listeners can hear this distinction slightly better than chance. Thus, instrumental data from one of the few languages in the world described as having nasalized fricatives confirms that the claimed sounds are not made by producing strong nasalization concurrently with clear frication noise. Furthermore, although speakers most often neutralize the nasalization, when they maintain it, they do so through a variety of phonetic mechanisms, even within a single language.

scholarly journals Evaluation of IMERG Level-3 Products in Depicting the July to October Rainfall over Taiwan: Typhoon Versus Non-Typhoon

2021 ◽  
Vol 13 (4) ◽  
pp. 622
Author(s):  
Wan-Ru Huang ◽  
Pin-Yi Liu ◽  
Ya-Hui Chang ◽  
Cheng-An Lee
Keyword(s):  
Real Time ◽  
Tropical Rainfall Measuring Mission ◽  
Temporal Variations ◽  
Rain Gauge ◽  
Advantages And Disadvantages ◽  
Level 3 ◽  
Spatio Temporal ◽  
Typhoon Season ◽  
Early Late ◽  
Better Than

This study assesses the performance of satellite precipitation products (SPPs) from the latest version, V06B, Integrated Multi-satellitE Retrievals for Global Precipitation Mission (IMERG) Level-3 (including early, late, and final runs), in depicting the characteristics of typhoon season (July to October) rainfall over Taiwan within the period of 2000–2018. The early and late runs are near-real-time SPPs, while final run is post-real-time SPP adjusted by monthly rain gauge data. The latency of early, late, and final runs is approximately 4 h, 14 h, and 3.5 months, respectively, after the observation. Analyses focus on the seasonal mean, daily variation, and interannual variation of typhoon-related (TC) and non-typhoon-related (non-TC) rainfall. Using local rain-gauge observations as a reference for evaluation, our results show that all IMERG products capture the spatio-temporal variations of TC rainfall better than those of non-TC rainfall. Among SPPs, the final run performs better than the late run, which is slightly better than the early run for most of the features assessed for both TC and non-TC rainfall. Despite these differences, all IMERG products outperform the frequently used Tropical Rainfall Measuring Mission 3B42 v7 (TRMM7) for the illustration of the spatio-temporal characteristics of TC rainfall in Taiwan. In contrast, for the non-TC rainfall, the final run performs notably better relative to TRMM7, while the early and late runs showed only slight improvement. These findings highlight the advantages and disadvantages of using IMERG products for studying or monitoring typhoon season rainfall in Taiwan.

No man is an island: The sustainability awareness effect of geography on hedonic fashion consumption and connection with nature – evidence from Galápagos and Hawaiʻi

2021 ◽  
Vol 8 (1-2) ◽  
pp. 205-221
Author(s):  
Attila Pohlmann
Keyword(s):  
United States ◽  
Supply Chain ◽  
Natural Environment ◽  
The United States ◽  
Santa Cruz ◽  
Galapagos Island ◽  
Fast Fashion ◽  
Trade Off ◽  
Visual Evidence ◽  
Connection With Nature

The demand for ever-accelerating fast fashion is unprecedented, while its supply chain burdens environmental systems. Hedonic fashion consumption is generally unfettered by sustainability concerns, but evidence suggests that island geographies–with dense boundaries between the built and the natural environment–have a heightening effect on eco-consciousness. A framework based on the contemporary condition of hyperconsumption is proposed: island geography heightens sustainability awareness; consequently, fashion consumers located on islands trade-off perceived hedonic benefits of fashion consumption against perceived moral benefits of connection with nature. The framework is supported by visual evidence collected on the Galápagos island Santa Cruz, indicating that male fashion consumers express connection with nature by means of tattoos, slogans on clothing and choice of eco-friendly materials. Quantitative tests with survey data from the United States and Ecuador show that residents in Hawaiʻi and the Galápagos have higher levels of connection with nature compared to residents on the associated continental areas. This effect is mediated by decreased perceived rewards of hedonic fashion consumption, but the effect is overall weaker in Ecuador compared to the United States due to differences in purchasing power and attitudes towards consumerism. Because of the stereotype that eco-friendly is unmanly, men are generally less likely to embrace environmentally friendly products and the findings of this research point to avenues to overcome this barrier.

COMPUTING THE RUPTURE DEGREE IN COMPOSITE GRAPHS

2010 ◽  
Vol 21 (03) ◽  
pp. 311-319 ◽  
Author(s):  
AYSUN AYTAC ◽  
ZEYNEP NIHAN ODABAS
Keyword(s):  
Complete Graph ◽  
Connected Graph ◽  
The Other ◽  
Trade Off ◽  
Number Of Components ◽  
Np Complete ◽  
Work Done ◽  
Better Than

The rupture degree of an incomplete connected graph G is defined by [Formula: see text] where w(G - S) is the number of components of G - S and m(G - S) is the order of a largest component of G - S. For the complete graph Kn, rupture degree is defined as 1 - n. This parameter can be used to measure the vulnerability of a graph. Rupture degree can reflect the vulnerability of graphs better than or independent of the other parameters. To some extent, it represents a trade-off between the amount of work done to damage the network and how badly the network is damaged. Computing the rupture degree of a graph is NP-complete. In this paper, we give formulas for the rupture degree of composition of some special graphs and we consider the relationships between the rupture degree and other vulnerability parameters.

scholarly journals Upscaling with the dynamic two-layer classification concept (D2C): TreeMig-2L, an efficient implementation of the forest-landscape model TreeMig

2015 ◽  
Vol 8 (7) ◽  
pp. 5535-5575
Author(s):  
J. E. M. S. Nabel
Keyword(s):  
Vegetation Dynamics ◽  
Climatic Changes ◽  
Discrete Models ◽  
Similarity Criteria ◽  
Forest Landscape ◽  
Two Dimensional ◽  
Landscape Model ◽  
Trade Off ◽  
Forest Landscape Model ◽  
Spatio Temporal

Abstract. Models used to investigate impacts of climatic changes on spatio-temporal vegetation dynamics need to balance required accuracy with computational feasibility. To enhance the computational efficiency of these models, upscaling methods are required that maintain key fine-scale processes influencing vegetation dynamics. In this paper, an adjustable method – the dynamic two-layer classification concept (D2C) – for the upscaling of time- and space-discrete models is presented. D2C aims to separate potentially repetitive calculations from those specific to single grid cells. The underlying idea is to extract processes that do not require information about a grid cell's neighbourhood to a reduced-size non-spatial layer, which is dynamically coupled to the original two-dimensional layer. The size of the non-spatial layer is thereby adaptive and depends on dynamic classifications according to pre-specified similarity criteria. I present how D2C can be used in a model implementation on the example of TreeMig-2L, a new, efficient version of the intermediate-complexity forest-landscape model TreeMig. To discuss the trade-off between computational expenses and accuracy, as well as the applicability of D2C, I compare different model stages of TreeMig-2L via simulations of two different application scenarios. This comparison of different model stages demonstrates that applying D2C can strongly reduce computational expenses of processes calculated on the new non-spatial layer. D2C is thus a valuable upscaling method for models and applications in which processes requiring information about the neighbourhood constitute the minor share of the overall computational expenses.

Sign in / Sign up

Export Citation Format

Share Document