scholarly journals GSM: Graph Similarity Model for Multi-Object Tracking

2020 ◽  
Author(s):  
Qiankun Liu ◽  
Qi Chu ◽  
Bin Liu ◽  
Nenghai Yu
Keyword(s):  
Object Tracking ◽  
Graph Matching ◽  
Feature Representation ◽  
Experimental Results ◽  
Graph Representation ◽  
Individual Object ◽  
Graph Similarity ◽  
Similarity Model ◽  
Similar Appearance ◽  
The Impact

The popular tracking-by-detection paradigm for multi-object tracking (MOT) focuses on solving data association problem, of which a robust similarity model lies in the heart. Most previous works make effort to improve feature representation for individual object while leaving the relations among objects less explored, which may be problematic in some complex scenarios. In this paper, we focus on leveraging the relations among objects to improve robustness of the similarity model. To this end, we propose a novel graph representation that takes both the feature of individual object and the relations among objects into consideration. Besides, a graph matching module is specially designed for the proposed graph representation to alleviate the impact of unreliable relations. With the help of the graph representation and the graph matching module, the proposed graph similarity model, named GSM, is more robust to the occlusion and the targets sharing similar appearance. We conduct extensive experiments on challenging MOT benchmarks and the experimental results demonstrate the effectiveness of the proposed method.

Occlusion Handling in Object Detection

2012 ◽  
pp. 61-74 ◽  
Author(s):  
Farjana Z. Eishita ◽  
Ashfaqur Rahman ◽  
Salahuddin A. Azad ◽  
Akhlaqur Rahman
Keyword(s):  
Object Detection ◽  
Object Tracking ◽  
Simple Algorithm ◽  
Texture Features ◽  
Image Plane ◽  
Experimental Results ◽  
Shape Information ◽  
Individual Object ◽  
Additional Information ◽  
Tracking Objects

Object tracking is a process that follows an object through consecutive frames of images to determine the object’s movement relative other objects of those frames. In other words, tracking is the problem of estimating the trajectory of an object in the image plane as it moves around a scene. This chapter presents research that deals with the problem of tracking objects when they are occluded. An object can be partially or fully occluded. Depending on the tracking domain, a tracker can deal with partial and full object occlusions using features such as colour and texture. But sometimes it fails to detect the objects after occlusion. The shape feature of an individual object can provide additional information while combined with colour and texture features. It has been observed that with the same colour and texture if two object’s shape information is taken then these two objects can be detected after the occlusion has occurred. From this observation, a new and a very simple algorithm is presented in this chapter, which is able to track objects after occlusion even if the colour and textures are the same. Some experimental results are shown along with several case studies to compare the effectiveness of the shape features against colour and texture features.

The Impact of Cash Mobs in the Vote with the Wallet Game: Experimental Results

2016 ◽  
Author(s):  
Leonardo Becchetti ◽  
Maurizio Fiaschetti ◽  
Francesco Salustri
Keyword(s):  
Experimental Results ◽  
The Impact

scholarly journals Experimental study of the supercritical CO 2 diffusion coefficient in porous media under reservoir conditions

2019 ◽  
Vol 6 (6) ◽  
pp. 181902 ◽  
Author(s):  
Junchen Lv ◽  
Yuan Chi ◽  
Changzhong Zhao ◽  
Yi Zhang ◽  
Hailin Mu
Keyword(s):  
Porous Media ◽  
Diffusion Coefficient ◽  
Oil Recovery ◽  
Carbon Capture ◽  
Carbon Capture And Storage ◽  
Elevated Pressure ◽  
Experimental Results ◽  
Saturated Porous Media ◽  
Reservoir Conditions ◽  
The Impact

Reliable measurement of the CO 2 diffusion coefficient in consolidated oil-saturated porous media is critical for the design and performance of CO 2 -enhanced oil recovery (EOR) and carbon capture and storage (CCS) projects. A thorough experimental investigation of the supercritical CO 2 diffusion in n -decane-saturated Berea cores with permeabilities of 50 and 100 mD was conducted in this study at elevated pressure (10–25 MPa) and temperature (333.15–373.15 K), which simulated actual reservoir conditions. The supercritical CO 2 diffusion coefficients in the Berea cores were calculated by a model appropriate for diffusion in porous media based on Fick's Law. The results show that the supercritical CO 2 diffusion coefficient increases as the pressure, temperature and permeability increase. The supercritical CO 2 diffusion coefficient first increases slowly at 10 MPa and then grows significantly with increasing pressure. The impact of the pressure decreases at elevated temperature. The effect of permeability remains steady despite the temperature change during the experiments. The effect of gas state and porous media on the supercritical CO 2 diffusion coefficient was further discussed by comparing the results of this study with previous study. Based on the experimental results, an empirical correlation for supercritical CO 2 diffusion coefficient in n -decane-saturated porous media was developed. The experimental results contribute to the study of supercritical CO 2 diffusion in compact porous media.

scholarly journals Offensive keyword extraction based on the attention mechanism of BERT and the eigenvector centrality using a graph representation

2021 ◽  
Author(s):  
Gretel Liz De la Peña Sarracén ◽  
Paolo Rosso
Keyword(s):  
Language Model ◽  
Hybrid Approach ◽  
Attention Mechanism ◽  
Graph Representation ◽  
Keyword Extraction ◽  
Eigenvector Centrality ◽  
Model Learning ◽  
Harmful Content ◽  
The Impact ◽  
Offensive Language

AbstractThe proliferation of harmful content on social media affects a large part of the user community. Therefore, several approaches have emerged to control this phenomenon automatically. However, this is still a quite challenging task. In this paper, we explore the offensive language as a particular case of harmful content and focus our study in the analysis of keywords in available datasets composed of offensive tweets. Thus, we aim to identify relevant words in those datasets and analyze how they can affect model learning. For keyword extraction, we propose an unsupervised hybrid approach which combines the multi-head self-attention of BERT and a reasoning on a word graph. The attention mechanism allows to capture relationships among words in a context, while a language model is learned. Then, the relationships are used to generate a graph from what we identify the most relevant words by using the eigenvector centrality. Experiments were performed by means of two mechanisms. On the one hand, we used an information retrieval system to evaluate the impact of the keywords in recovering offensive tweets from a dataset. On the other hand, we evaluated a keyword-based model for offensive language detection. Results highlight some points to consider when training models with available datasets.

scholarly journals Three doors anomaly, “should I stay, or should I go”: an artefactual field experiment

2021 ◽  
Author(s):  
Andrea Morone ◽  
Rocco Caferra ◽  
Alessia Casamassima ◽  
Alessandro Cascavilla ◽  
Paola Tiranzoni
Keyword(s):  
Field Experiment ◽  
Experimental Evidence ◽  
Bayesian Updating ◽  
Anomalous Behavior ◽  
Status Quo ◽  
Experimental Results ◽  
Illusion Of Control ◽  
Optimal Decision ◽  
Status Quo Bias ◽  
The Impact

AbstractThis work aims to identify and quantify the biases behind the anomalous behavior of people when they deal with the Three Doors dilemma, which is a really simple but counterintuitive game. Carrying out an artefactual field experiment and proposing eight different treatments to isolate the anomalies, we provide new interesting experimental evidence on the reasons why subjects fail to take the optimal decision. According to the experimental results, we are able to quantify the size and the impact of three main biases that explain the anomalous behavior of participants: Bayesian updating, illusion of control and status quo bias.

scholarly journals Research of Personalized Recommendation Technology Based on Knowledge Graphs

2021 ◽  
Vol 11 (15) ◽  
pp. 7104
Author(s):  
Xu Yang ◽  
Ziyi Huan ◽  
Yisong Zhai ◽  
Ting Lin
Keyword(s):  
Neural Network ◽  
Hot Spot ◽  
Experimental Results ◽  
Graph Representation ◽  
Personalized Recommendation ◽  
Knowledge Graph ◽  
Learning Technology ◽  
Data Set ◽  
Model Based ◽  
Knowledge Graphs

Nowadays, personalized recommendation based on knowledge graphs has become a hot spot for researchers due to its good recommendation effect. In this paper, we researched personalized recommendation based on knowledge graphs. First of all, we study the knowledge graphs’ construction method and complete the construction of the movie knowledge graphs. Furthermore, we use Neo4j graph database to store the movie data and vividly display it. Then, the classical translation model TransE algorithm in knowledge graph representation learning technology is studied in this paper, and we improved the algorithm through a cross-training method by using the information of the neighboring feature structures of the entities in the knowledge graph. Furthermore, the negative sampling process of TransE algorithm is improved. The experimental results show that the improved TransE model can more accurately vectorize entities and relations. Finally, this paper constructs a recommendation model by combining knowledge graphs with ranking learning and neural network. We propose the Bayesian personalized recommendation model based on knowledge graphs (KG-BPR) and the neural network recommendation model based on knowledge graphs(KG-NN). The semantic information of entities and relations in knowledge graphs is embedded into vector space by using improved TransE method, and we compare the results. The item entity vectors containing external knowledge information are integrated into the BPR model and neural network, respectively, which make up for the lack of knowledge information of the item itself. Finally, the experimental analysis is carried out on MovieLens-1M data set. The experimental results show that the two recommendation models proposed in this paper can effectively improve the accuracy, recall, F1 value and MAP value of recommendation.

Interpretation of reflexive anaphora in second language VP-ellipsis: Relevance Theory and paradigms of explanation

2004 ◽  
Vol 20 (3) ◽  
pp. 256-280 ◽  
Author(s):  
Xosé Rosales Sequeiros
Keyword(s):  
Second Language ◽  
Relevance Theory ◽  
Experimental Results ◽  
Comprehension Strategies ◽  
Processing Cost ◽  
L2 Learners ◽  
Vp Ellipsis ◽  
Processing Effort ◽  
Maximal Relevance ◽  
The Impact

This article explores second language (L2) learners’ interpretation of reflexive anaphora in VP-Ellipsis by critiquing the work of Ying (2003), who applies Relevance Theory to explain elliptical anaphora. It argues against four claims made in his analysis: that L2 learners apply maximal relevance in anaphoric interpretation; that a procedural account of the impact of referential sentences on VP-ellipsis disambiguation is appropriate; that an account of anaphoric interpretation preferences should be based on processing cost; and that differences in experimental results between intermediate and advanced L2 learners are due to the use of different comprehension strategies (see Sperber, 1994). Instead, it argues: that it is not maximal but rather optimal relevance that is at work; that the key in disambiguating anaphora in VP-elliptical sentences is the achievement of an optimally relevant interpretation; that the role of contextual assumptions in anaphora resolution is to enable L2 learners to derive enough contextual effects to make it worth their effort and, in doing so, identifying (as a side effect) what they take to have been the intended referent; and that what is crucial in the use of comprehension strategies is not processing effort, but rather consistency with the second principle of relevance. Overall, all these factors provide the basis for an alternative and more comprehensive explanation of the experimental results discussed by Ying.

Study on the Dyeing Behavior of Wool Fiber Treated with Nano-SiO2/Ag Antibacterial Agent

2011 ◽  
Vol 332-334 ◽  
pp. 27-30 ◽  
Author(s):  
Mei Niu ◽  
Zi Lu Wu ◽  
Jin Ming Dai ◽  
Wen Sheng Hou ◽  
Sheng Shi ◽  
...  
Keyword(s):  
Low Temperature ◽  
Antibacterial Agent ◽  
Important Influence ◽  
Wool Fiber ◽  
Experimental Results ◽  
Ultrasonic Frequency ◽  
Acid Dyes ◽  
Nano Sio2 ◽  
Dyeing Process ◽  
The Impact

Wool fiber was firstly pretreated by nano-SiO2/Ag antibacterial agent, and then dyed with an acid dyes at low temperature by ultrasonic dyeing. Many factors had an important influence on the dye ability and the antibacterial behavior during the dyeing process of antibacterial wool fiber. The experimental results indicate that the dye-takeup rates of antibacterial wool fiber were enhanced with the increase of the concentration of nano-SiO2/Ag, the dyeing temperature, the dyeing time and the ultrasonic frequency (less than 60Hz). However, the antibacterial ratios of wool fiber were declined in the impact of these factors other than the concentration of antibacterial agent.

Sign in / Sign up

Export Citation Format

Share Document