scholarly journals FASSD-Net Model for Person Semantic Segmentation

Electronics ◽  
2021 ◽  
Vol 10 (12) ◽  
pp. 1393
Author(s):  
Luis Brandon Garcia-Ortiz ◽  
Jose Portillo-Portillo ◽  
Aldo Hernandez-Suarez ◽  
Jesus Olivares-Mercado ◽  
Gabriel Sanchez-Perez ◽  
...  
Keyword(s):  
Semantic Segmentation ◽  
Human Interaction ◽  
Video Sequences ◽  
Qualitative And Quantitative ◽  
High Resolution Images ◽  
Quantitative Manner ◽  
High Level ◽  
Time Segmentation ◽  
Human Silhouette ◽  
Silhouette Extraction

This paper proposes the use of the FASSD-Net model for semantic segmentation of human silhouettes, these silhouettes can later be used in various applications that require specific characteristics of human interaction observed in video sequences for the understanding of human activities or for human identification. These applications are classified as high-level task semantic understanding. Since semantic segmentation is presented as one solution for human silhouette extraction, it is concluded that convolutional neural networks (CNN) have a clear advantage over traditional methods for computer vision, based on their ability to learn the representations of appropriate characteristics for the task of segmentation. In this work, the FASSD-Net model is used as a novel proposal that promises real-time segmentation in high-resolution images exceeding 20 FPS. To evaluate the proposed scheme, we use the Cityscapes database, which consists of sundry scenarios that represent human interaction with its environment (these scenarios show the semantic segmentation of people, difficult to solve, that favors the evaluation of our proposal), To adapt the FASSD-Net model to human silhouette semantic segmentation, the indexes of the 19 classes traditionally proposed for Cityscapes were modified, leaving only two labels: One for the class of interest labeled as person and one for the background. The Cityscapes database includes the category “human” composed for “rider” and “person” classes, in which the rider class contains incomplete human silhouettes due to self-occlusions for the activity or transport used. For this reason, we only train the model using the person class rather than human category. The implementation of the FASSD-Net model with only two classes shows promising results in both a qualitative and quantitative manner for the segmentation of human silhouettes.

A Fast-RCNN Implementation for Human Silhouette Detection in Video Sequences

2020 ◽  
Author(s):  
Luis Brandon Garcia-Ortiz ◽  
Gabriel Sanchez-Perez ◽  
Aldo Hernandez-Suarez ◽  
Jesus Olivares-Mercado ◽  
Hector Manuel Perez-Meana ◽  
...  
Keyword(s):  
Fall Detection ◽  
Human Interaction ◽  
Video Sequences ◽  
Surveillance Systems ◽  
Research Areas ◽  
The Common ◽  
Behavioural Biometrics ◽  
Silhouette Detection ◽  
Human Silhouette ◽  
Silhouette Extraction

The intention of this article is to implement a system of detection and segmentation of human silhouettes, the above mentioned tasks present a great challenge in security topics and innovation, in the last years and mainly on automated video surveillance systems, which require understanding the presence and human interaction in video sequences, e.g. Human Computer Interaction (HCI), Human Behaviour comprehension, Human fall detection, among others, but the most important is behavioural biometrics, this paper tackles the common step in these research areas: the Human silhouette extraction through the bounding box. To evaluate the proposed system, standardized databases where used and also proper videos are obtained trying to emulate real-world scenarios, where the quality and the distance are factors that have demonstrated challenges for the detection with computer vision and machine learning.

Pollen morphology of some Tamarix (Tamaricaceae) species of Iran

2020 ◽  
Vol 62 (1-2) ◽  
pp. 151-161
Author(s):  
T. Shagholi ◽  
M. Keshavarzi ◽  
M. Sheidai
Keyword(s):  
North Africa ◽  
Pollen Grains ◽  
Ecological Niches ◽  
Multivariate Statistical Methods ◽  
Multivariate Statistical ◽  
Qualitative And Quantitative ◽  
Petal Length ◽  
Electron Microscopes ◽  
High Level ◽  
Scanning Electron Microscopes

Tamarix L. (Tamaricaceae) is a halophytic shrub in different parts of Asia and North Africa. Taxonomy and species limitation of Tamarix is very complex. This genus has three sections as Tamarix, Oligadenia, and Polyadenia, which are mainly separated by petal length, the number of stamens, the shape of androecial disk and attachment of filament on the androecial disk. As there was no palynological data on pollen features of Tamarix species of Iran, in the present study 12 qualitative and quantitative pollen features were evaluated to find diagnostic ones. Pollen grains of 8 Tamarix species were collected from nature. Pollen grains were studied without any treatment. Measurements were based on at least 50 pollen grains per specimen. Light and scanning electron microscopes were used. Multivariate statistical methods were applied to clarify the species relationships based on pollen data. All species studied showed monad and tricolpate (except some individuals of T. androssowii). Some Tamarix species show a high level of variability, in response to ecological niches and phenotypic plasticity, which make Tamarix species separation much more difficult. Based on the results of the present study, pollen grains features are not in agreement with previous morphological and molecular genetics about the sectional distinction.

Splash formation by spherical drops

2001 ◽  
Vol 427 ◽  
pp. 73-105 ◽  
Author(s):  
LIOW JONG LENG
Keyword(s):  
High Resolution ◽  
Quantitative Data ◽  
Water Surface ◽  
High Speed ◽  
Scaling Laws ◽  
Capillary Wave ◽  
Water Drop ◽  
Qualitative And Quantitative ◽  
High Resolution Images ◽  
The Impact

The impact of a spherical water drop onto a water surface has been studied experimentally with the aid of a 35 mm drum camera giving high-resolution images that provided qualitative and quantitative data on the phenomena. Scaling laws for the time to reach maximum cavity sizes have been derived and provide a good fit to the experimental results. Transitions between the regimes for coalescence-only, the formation of a high-speed jet and bubble entrapment have been delineated. The high-speed jet was found to occur without bubble entrapment. This was caused by the rapid retraction of the trough formed by a capillary wave converging to the centre of the cavity base. The converging capillary wave has a profile similar to a Crapper wave. A plot showing the different regimes of cavity and impact drop behaviour in the Weber–Froude number-plane has been constructed for Fr and We less than 1000.

scholarly journals EXPLORING ALS AND DIM DATA FOR SEMANTIC SEGMENTATION USING CNNS

2018 ◽  
Vol XLII-1 ◽  
pp. 347-354 ◽  
Author(s):  
F. Politz ◽  
M. Sester
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Semantic Segmentation ◽  
Point Clouds ◽  
Good Alternative ◽  
Aerial Images ◽  
Learning Approaches ◽  
Advantages And Disadvantages ◽  
Sensing Applications ◽  
High Level

<p><strong>Abstract.</strong> Over the past years, the algorithms for dense image matching (DIM) to obtain point clouds from aerial images improved significantly. Consequently, DIM point clouds are now a good alternative to the established Airborne Laser Scanning (ALS) point clouds for remote sensing applications. In order to derive high-level applications such as digital terrain models or city models, each point within a point cloud must be assigned a class label. Usually, ALS and DIM are labelled with different classifiers due to their varying characteristics. In this work, we explore both point cloud types in a fully convolutional encoder-decoder network, which learns to classify ALS as well as DIM point clouds. As input, we project the point clouds onto a 2D image raster plane and calculate the minimal, average and maximal height values for each raster cell. The network then differentiates between the classes ground, non-ground, building and no data. We test our network in six training setups using only one point cloud type, both point clouds as well as several transfer-learning approaches. We quantitatively and qualitatively compare all results and discuss the advantages and disadvantages of all setups. The best network achieves an overall accuracy of 96<span class="thinspace"></span>% in an ALS and 83<span class="thinspace"></span>% in a DIM test set.</p>

scholarly journals HYBRID ACQUISITION OF HIGH QUALITY TRAINING DATA FOR SEMANTIC SEGMENTATION OF 3D POINT CLOUDS USING CROWD-BASED ACTIVE LEARNING

2020 ◽  
Vol V-2-2020 ◽  
pp. 501-508
Author(s):  
M. Kölle ◽  
V. Walter ◽  
S. Schmohl ◽  
U. Soergel
Keyword(s):  
Active Learning ◽  
Semantic Segmentation ◽  
Point Clouds ◽  
Human Interaction ◽  
Training Data ◽  
Semantic Interpretation ◽  
Training Dataset ◽  
3D Point Clouds ◽  
Semantic Labeling ◽  
3D Cnn

Abstract. Automated semantic interpretation of 3D point clouds is crucial for many tasks in the domain of geospatial data analysis. For this purpose, labeled training data is required, which has often to be provided manually by experts. One approach to minimize effort in terms of costs of human interaction is Active Learning (AL). The aim is to process only the subset of an unlabeled dataset that is particularly helpful with respect to class separation. Here a machine identifies informative instances which are then labeled by humans, thereby increasing the performance of the machine. In order to completely avoid involvement of an expert, this time-consuming annotation can be resolved via crowdsourcing. Therefore, we propose an approach combining AL with paid crowdsourcing. Although incorporating human interaction, our method can run fully automatically, so that only an unlabeled dataset and a fixed financial budget for the payment of the crowdworkers need to be provided. We conduct multiple iteration steps of the AL process on the ISPRS Vaihingen 3D Semantic Labeling benchmark dataset (V3D) and especially evaluate the performance of the crowd when labeling 3D points. We prove our concept by using labels derived from our crowd-based AL method for classifying the test dataset. The analysis outlines that by labeling only 0:4% of the training dataset by the crowd and spending less than 145 $, both our trained Random Forest and sparse 3D CNN classifier differ in Overall Accuracy by less than 3 percentage points compared to the same classifiers trained on the complete V3D training set.

scholarly journals The Emergence of Nature in Contemporary Iranian Architecture

2015 ◽  
Vol 10 (Special-Issue1) ◽  
pp. 144-156
Author(s):  
Mahmood Feizabadi ◽  
Mohammadjavad Mahdavinejad ◽  
Seyyed Mirhosseini
Keyword(s):  
Literature Review ◽  
Case Studies ◽  
Evaluation Criteria ◽  
Public Works ◽  
Basic Question ◽  
Analytic Method ◽  
Public Buildings ◽  
Qualitative And Quantitative ◽  
Quantitative Manner ◽  
Contemporary Architecture

In this study, we discuss ways of affecting nature on contemporary architecture and utilise them to survey the naturalism of case studies of Iranian architecture. The basic question in this study is: 'how ways of utilising nature have influenced on contemporary public works of Iran?' Descriptive-analytic method is used to achieve the results. The literature review was done by using archival methods, then the ways of affecting nature on contemporary architecture were listed as an evaluation criteria. Next, characteristics of sample projects were analyzed by using surveying methods, and their effects were submitted in qualitative and quantitative manner. The results of the study showed that some ways of affecting nature include scenery, material and conceptual have had the most usage in contemporary public buildings of Iran, and some others include spatial, functional and formal have been overlooked.

scholarly journals Qualitative and Quantitative Integrated Modeling for Stochastic Simulation and Optimization

2013 ◽  
Vol 2013 ◽  
pp. 1-12 ◽  
Author(s):  
Xuefeng Yan ◽  
Yong Zhou ◽  
Yan Wen ◽  
Xudong Chai
Keyword(s):  
Hierarchical Model ◽  
Description Logic ◽  
Expert Knowledge ◽  
Uncertain Reasoning ◽  
Model Structure ◽  
Meta Model ◽  
Simulation And Optimization ◽  
Qualitative And Quantitative ◽  
Level Model ◽  
High Level

The simulation and optimization of an actual physics system are usually constructed based on the stochastic models, which have both qualitative and quantitative characteristics inherently. Most modeling specifications and frameworks find it difficult to describe the qualitative model directly. In order to deal with the expert knowledge, uncertain reasoning, and other qualitative information, a qualitative and quantitative combined modeling specification was proposed based on a hierarchical model structure framework. The new modeling approach is based on a hierarchical model structure which includes the meta-meta model, the meta-model and the high-level model. A description logic system is defined for formal definition and verification of the new modeling specification. A stochastic defense simulation was developed to illustrate how to model the system and optimize the result. The result shows that the proposed method can describe the complex system more comprehensively, and the survival probability of the target is higher by introducing qualitative models into quantitative simulation.

Y-Net: Dual-branch Joint Network for Semantic Segmentation

2021 ◽  
Vol 17 (4) ◽  
pp. 1-22
Author(s):  
Yizhen Chen ◽  
Haifeng Hu
Keyword(s):  
Feature Vector ◽  
State Of The Art ◽  
Computational Cost ◽  
Receptive Fields ◽  
Semantic Segmentation ◽  
Global Context ◽  
Multi Level ◽  
The One ◽  
Public Datasets ◽  
High Level

Most existing segmentation networks are built upon a “ U -shaped” encoder–decoder structure, where the multi-level features extracted by the encoder are gradually aggregated by the decoder. Although this structure has been proven to be effective in improving segmentation performance, there are two main drawbacks. On the one hand, the introduction of low-level features brings a significant increase in calculations without an obvious performance gain. On the other hand, general strategies of feature aggregation such as addition and concatenation fuse features without considering the usefulness of each feature vector, which mixes the useful information with massive noises. In this article, we abandon the traditional “ U -shaped” architecture and propose Y-Net, a dual-branch joint network for accurate semantic segmentation. Specifically, it only aggregates the high-level features with low-resolution and utilizes the global context guidance generated by the first branch to refine the second branch. The dual branches are effectively connected through a Semantic Enhancing Module, which can be regarded as the combination of spatial attention and channel attention. We also design a novel Channel-Selective Decoder (CSD) to adaptively integrate features from different receptive fields by assigning specific channelwise weights, where the weights are input-dependent. Our Y-Net is capable of breaking through the limit of singe-branch network and attaining higher performance with less computational cost than “ U -shaped” structure. The proposed CSD can better integrate useful information and suppress interference noises. Comprehensive experiments are carried out on three public datasets to evaluate the effectiveness of our method. Eventually, our Y-Net achieves state-of-the-art performance on PASCAL VOC 2012, PASCAL Person-Part, and ADE20K dataset without pre-training on extra datasets.

Sign in / Sign up

Export Citation Format

Share Document