2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ)最新文献

Pasture Fence Line Detection in UAV Videos 无人机视频中的牧场围栏线检测

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653285

Farah Sarwar, Anthony Griffin, P. Chong, T. Pasang

引用次数: 1

Denomination-based Quasi-isotropic Remeshing using Curvature Map and Voronoi Tessellation 基于曲率映射和Voronoi曲面细分的准各向同性网格划分

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653252

Ashutosh Soni, Preetam Chayan Chatterjee, Piyush Kanti Bhunre, Partha Bhowmick

{"title":"Denomination-based Quasi-isotropic Remeshing using Curvature Map and Voronoi Tessellation","authors":"Ashutosh Soni, Preetam Chayan Chatterjee, Piyush Kanti Bhunre, Partha Bhowmick","doi":"10.1109/IVCNZ54163.2021.9653252","DOIUrl":"https://doi.org/10.1109/IVCNZ54163.2021.9653252","url":null,"abstract":"We propose a novel algorithm for quasi-isotropic remeshing when edge-length denomination is given as input. It is designed by the prudence of fitting triangles on the surface based on curvature—higher the curvature, lower will be the triangle size (as per the given denomination), and vice versa. By virtue of this, we have a better control of the remeshing quality, measured in terms of different parameters such as Hausdorff error, average vertex degree, and the like. The proposed pipeline comprises mesh voxelization, discrete curvature estimation, centroidal Voronoi tessellation based on curvature map and energy optimization, and finally the conventional step of Delaunay triangulation. Upon testing on various datasets, it is found to exhibit encouraging results with isotropic triangles getting fitted in regions of uniform curvature, while some occasional anisotropy occurring across regions of changing curvature. We have furnished related results and statistics to adjudge its merit, and have pointed out its further scope in mesh remodeling.","PeriodicalId":413244,"journal":{"name":"2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ)","volume":"37 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114033531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Robust human instance segmentation in a challenging forest environment 具有挑战性的森林环境中鲁棒的人类实例分割

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653172

K. Pahalawatta, J. Fourie, J. Potgieter, Heath Ascot-Evans, Armin Werner

引用次数: 0

Pointing-Based Calibration of a Pointing Interface for Public Displays 公共显示器指向接口的基于指向的校准

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653203

S. Mills

引用次数: 0

Autonomous Detection of Facial Tics in Video Recordings 录像中面部抽搐的自动检测

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653306

Jack van Heugten Breurkes, R. Green

{"title":"Autonomous Detection of Facial Tics in Video Recordings","authors":"Jack van Heugten Breurkes, R. Green","doi":"10.1109/IVCNZ54163.2021.9653306","DOIUrl":"https://doi.org/10.1109/IVCNZ54163.2021.9653306","url":null,"abstract":"Facial tics are abrupt and involuntary facial movements that are typically expressed for less than one second. This paper proposes a method to autonomously detect facial tics in video recordings to help medical professionals diagnose and monitor tic disorders in patients.Faces are detected in the video feed using a set of Histogram of Oriented Gradients filters and a linear Support Vector Machine. Facial movements are monitored by tracking the area and the horizontal and vertical coordinates of each of the patient’s facial features. Facial features are detected by calculating the convex hulls of points on the face known as facial landmarks. Facial landmark detection is implemented using a cascade of regressor functions that are trained to find the positions of 68 facial landmarks. The coordinates and area of each facial feature are analysed in separate data streams. When the value of a point in a data stream differs significantly from the average value of the surrounding points, a significant movement is detected. Short duration significant movements are labelled as facial tics. Long duration significant movements are labelled as long duration movements.The proposed method was able to achieve 90% accuracy in controlled conditions. 66% of long duration significant movements were incorrectly labelled as facial tics, giving a 66% false positive rate. The proposed method improves on past research by analysing multiple facial features and by providing graphical information on the characteristics of each facial movement.","PeriodicalId":413244,"journal":{"name":"2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127696955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Relative Camera Rotation from a Single Oriented Correspondence 从单一方向对应的相对相机旋转

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653298

S. Mills

引用次数: 1

[IVCNZ 2021 Front cover] [IVCNZ 2021封面]

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/ivcnz54163.2021.9653278

引用次数: 0

Unsupervised Learning of Affinity for Image Segmentation: An Inpainting based Approach 图像分割中亲和力的无监督学习:一种基于Inpainting的方法

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653321

Swati Bhugra, V. Kaushik, I. Mateos, S. Chaudhury, Brejesh Lall

{"title":"Unsupervised Learning of Affinity for Image Segmentation: An Inpainting based Approach","authors":"Swati Bhugra, V. Kaushik, I. Mateos, S. Chaudhury, Brejesh Lall","doi":"10.1109/IVCNZ54163.2021.9653321","DOIUrl":"https://doi.org/10.1109/IVCNZ54163.2021.9653321","url":null,"abstract":"Unsupervised image segmentation is a fundamental problem that aims at clustering pixels into regions satisfying homogeneity (w.r.t color or texture), with no pixel annotations specified as ground truth. This paper proposes a novel unsupervised graph based segmentation framework based on a novel affinity matrix. The main hypothesis is that a convolutional neural network (CNN) trained to generate the contents of arbitrary missing image regions can provide powerful grouping cues among pixels in a given image. We encode this intuition with the generation of activation maps from state-of-the-art inpainting network, where the missing region corresponds to a superpixel of the input image. The activation weights are obtained for each pixel of every superpixel region and are utilised in graph construction over these superpixels. In contrast to the traditional graph based methods, that are primarily dependent on integrating local grouping cues, we propose to learn affinities based on high feature descriptiveness of CNNs, thus enabling reliable graph building for segmentation. Experimental results on the Berkeley segmentation database demonstrates the effectiveness of the proposed framework.","PeriodicalId":413244,"journal":{"name":"2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124848684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Small mammals and bird detection using IoT devices 使用物联网设备检测小型哺乳动物和鸟类

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653430

Kyuwon Shim, A. Barczak, N. Reyes, N. Ahmed

{"title":"Small mammals and bird detection using IoT devices","authors":"Kyuwon Shim, A. Barczak, N. Reyes, N. Ahmed","doi":"10.1109/IVCNZ54163.2021.9653430","DOIUrl":"https://doi.org/10.1109/IVCNZ54163.2021.9653430","url":null,"abstract":"This article describes the application of computer vision to automatically detect pests in protected forests. One of the most threatening pests in NZ are rodents (rats, mice, possums etc). The objective of the application is to allow the deployment of Internet of Things (IoT) devices capable of detecting rodents while differentiating them from birds. As IoT devices have limited memory and CPU capabilities, it is important to find computer vision methods that are efficient and lightweight. Two main methods were employed for this work to test their suitability for this application. The first method is based on getting a profile of moving objects and classify them using Fourier descriptors (FD) and classifiers. The second method uses YOLO (You Only Look Once) to classify birds versus rodents. In order to increase the number of images for training YOLO, a semi-automatic labelling system was created, using the FD method to segment images of birds and rodents. The final accuracy rate using FD reached 83 percent with random forests, and YOLO reached 97 percent. The FD method can outperform YOLO in terms of speed when using a Raspberry Pi, achieving more than 6 frames per second.","PeriodicalId":413244,"journal":{"name":"2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124969270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Deep Convolutional Neural Networks for Fish Weight Prediction from Images 基于图像的深度卷积神经网络预测鱼类体重

2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ) Pub Date : 2021-12-09 DOI: 10.1109/IVCNZ54163.2021.9653412

Yunhan Yang, Bing Xue, L. Jesson, Matthew Wylie, Mengjie Zhang, M. Wellenreuther

{"title":"Deep Convolutional Neural Networks for Fish Weight Prediction from Images","authors":"Yunhan Yang, Bing Xue, L. Jesson, Matthew Wylie, Mengjie Zhang, M. Wellenreuther","doi":"10.1109/IVCNZ54163.2021.9653412","DOIUrl":"https://doi.org/10.1109/IVCNZ54163.2021.9653412","url":null,"abstract":"Fish weight is an important performance trait in aquaculture, conservation, fisheries science and management since weight relates to the growth of individual fish in a particular environment. A power regression model is commonly used to explain the relationship between fish weight and length. However, this requires costly measurements of fish length. The present study applies machine learning techniques to predict fish weight from fish images, bypassing the length measurement step. In this study, we validate the feasibility of predicting fish weight from images directly. We use a convolutional neural networks (CNNs) based approach to predict fish weight from images by building regression models. The deep CNNs architecture VGG-11, ResNet-18 and DenseNet-121 are chosen to train the models. The fish images have different scales (length-pixel ratio) without including a ruler as a reference. The trained regressors of these three architectures reach R2 0.94, 0.95 and 0.96 on the test set. Our results support the feasibility of fish weight prediction with the CNNs model from images directly. The fish images look similar to humans, but CNNs regressors can detect the different fish weights. The CNNs regressors also can detect the fish images with different length-pixel ratios.","PeriodicalId":413244,"journal":{"name":"2021 36th International Conference on Image and Vision Computing New Zealand (IVCNZ)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127060987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1