A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
Anchor-based Nearest Class Mean Loss for Convolutional Neural Networks
[article]
2018
arXiv
pre-print
Preserved Fulltext
Most existing deep learning approaches, however, rely on convolutional neural networks (CNNs) for learning features, whose discriminant power is not explicitly enforced. ...
To this end, we introduce anchors, which are predefined vectors regarded as the centers for each class and fixed during training. ...
The proposed anchor-based nearest class mean loss differs from SoftMax in two aspects: 1) the deep feature dimensions for anchor-based nearest class mean loss are not limited, while the softmax loss requires ...
arXiv:1804.08087v1
fatcat:ellhx6humnb6ffby27sjduyhz4
Personalized Activity Recognition with Deep Triplet Embeddings
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We present an approach to personalized activity recognition based on deep embeddings derived from a fully convolutional neural network. ...
neural network classifier. ...
Impersonal Fully Convolutional Network (FCN) A fully convolutional neural network (FCN) was utilized as the baseline impersonal supervised learning model. ...
arXiv:2001.05517v1
fatcat:wppwc3ykf5clrhrmlu7lmsae74
Revealing the Unknown: Real-Time Recognition of Galápagos Snake Species Using Deep Learning
2020
Animals
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Using the deep learning and machine learning algorithms and libraries, we modified and successfully implemented four region-based convolutional neural network (R-CNN) architectures (models for image classification ...
Real-time identification of wildlife is an upcoming and promising tool for the preservation of wildlife. ...
R-CNN (region-based convolutional neural network) is widely used for object detection, whereas CNN (convolutional neural network) is used for image classification. ...
doi:10.3390/ani10050806
pmid:32384793
pmcid:PMC7278857
fatcat:65jf6ectozasrplz7piwu23xke
LittleYOLO-SPP: A Delicate Real-Time Vehicle Detection Algorithm
2020
Optik (Stuttgart)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2020 pre-print arXiv:2011.05940v1 fatcat:gbs655khbrdjjojoxkduriejsy
The Mean square error (MSE) and Generalized IoU (GIoU) loss function for bounding box regression is used to increase the performance of the network. ...
The network training includes vehicle-based classes from PASCAL VOC 2007,2012 and MS COCO 2014 datasets such as car, bus, and truck. ...
high accuracy, which makes the network suitable for high-end surveillance systems. However, the algorithm is still missed to detect some vehicles and made few labelling mistakes too. ...
doi:10.1016/j.ijleo.2020.165818
fatcat:ehm5xoedinbjzitxc7bocyli7m
Estimating and abstracting the 3D structure of bones using neural networks on X-ray (2D) images
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Our triplet loss-trained neural network selects the most closely matching 3D bone shape from a predefined set of shapes. ...
Additionally, our neural network can determine the identity of a bone based only on its X-ray image. ...
Data and software availability We declare that the programming code necessary to reproduce the findings of this paper (code for data pre-processing, training the Triplet neural network and kNN classifier ...
arXiv:2001.11499v1
fatcat:4cxhq3arzrbejmpwhtrw4admqe
Incremental Class Learning using Variational Autoencoders with Similarity Learning
[article]
2023
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Our research investigates catastrophic forgetting for four well-known similarity-based loss functions during incremental class learning. ...
Applications of neural networks have been extended to include similarity learning. ...
Acknowledgements We want to acknowledge the Nedbank Research and Innovation Chair for supporting this research. ...
arXiv:2110.01303v3
fatcat:g6s2hqa6tvgmhm62xtpxn3ru5q
LANDMARK-BASED VISUAL PLACE RECOGNITION
2021
International Journal of Engineering Applied Sciences and Technology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
By fine-tuning pre-trained convolutional neural network (CNN) and minimizing triplet loss, the triplet network can learn appropriate metrics so that most similar images can be retrieved through algorithms ...
for the k-nearest neighbor (KNN). ...
Triplet loss is a loss function of artificial neural networks where baseline (anchor) input is compared to positive (true) input and negative (false) feedback. ...
doi:10.33564/ijeast.2021.v05i11.035
fatcat:p7dwn5r4ivhcdko7n2nuin6fya
Multi-Person Tracking Based on Faster R-CNN and Deep Appearance Features
[chapter]
2019
Visual Object Tracking in the Deep Neural Networks Era [Working Title]
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Object detection accuracy has been increased by employing deep learning-based Faster region convolutional neural network (Faster R-CNN) algorithm. ...
This paper presents a novel algorithm for improved object detection followed by enhanced object tracking. ...
Faster R-CNN is a purely convolution neural network without any handcrafted features that employ fully convolution neural network (FCN) for region proposal. ...
doi:10.5772/intechopen.85215
fatcat:oc3kfowngfei5gmeh33uu4dwhy
A Target Detection Model Based on Improved Tiny-yolov3 Under the Environment of Mining Truck
2019
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
INDEX TERMS Convolutional neural network, real-time, residual network, target detection, tiny-yolov3. ...
The residual network structure based on convolutional neural network is added to the tiny-yolov3 structure, and the accuracy of obstacle detection is improved under the condition of real-time detection ...
The Region Based Convolutional Neural Network (RCNN) raises the accuracy of target detection to a new level. ...
doi:10.1109/access.2019.2928603
fatcat:lh4pn3p6onfvtlm6wlv5552cnq
Personalized Activity Recognition with Deep Triplet Embeddings
2022
Sensors
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
We present an approach to personalized activity recognition based on deep feature representation derived from a convolutional neural network (CNN). ...
We experiment with both categorical cross-entropy loss and triplet loss for training, and describe a novel loss function based on subject triplets. ...
In the work by He et al. triplets were sampled based on a hierarchical strategy in the application of fine-grained image classification, where a convolutional neural network was trained to extract low-level ...
doi:10.3390/s22145222
pmid:35890902
pmcid:PMC9324610
fatcat:ieguoibxzzh4netm4cr6drozba
Comparing Class-Aware and Pairwise Loss Functions for Deep Metric Learning in Wildlife Re-Identification
2021
Sensors
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Similarity learning using deep convolutional neural networks has been applied extensively in solving computer vision problems. ...
We demonstrate that no single neural network architecture combined with a loss function is best suited for all datasets, although VGG-11 may be the most robust first choice. ...
Acknowledgments: We would like to thank the Nedbank Research Chair for the support provided to us while we undertook this research. ...
doi:10.3390/s21186109
pmid:34577319
fatcat:y7d6k7nsb5d7hjpe76rp5mzddy
Semi-supervised Triplet Loss Based Learning of Ambient Audio Embeddings
2019
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Deep neural networks are particularly useful to learn relevant representations from data. ...
In this paper, we combine unsupervised and supervised triplet loss based learning into a semi-supervised representation learning approach. ...
State-of-the-art methods for audio tagging rely on deep neural networks (DNNs), such as convolutional neural networks (CNNs) [3, 4] , recurrent neural networks (RNNs) [5] , or a combination of both usually ...
doi:10.1109/icassp.2019.8683774
dblp:conf/icassp/TurpaultSV19
fatcat:f4pcyw36gbh2fd4sqwxi3jkkem
Metric Learning Based Convolutional Neural Network for Left-Right Brain Dominance Classification
2021
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
In this paper, we employ a series of EEG signal processing techniques and a state-of-the-art deep learning neural network namely Metric Learning Based Convolutional Neural Network (MLBCNN) to determine ...
INDEX TERMS Brain dominance, electroencephalogram, deep learning, metric learning, convolutional neural network. ...
METRIC LEARNING BASED CONVOLUTIONAL NEURAL NETWORK (MLBCNN) FOR CLASSIFICATION In this research, the deep neural network model employed for classification is a convolutional neural network (CNN) with 3 ...
doi:10.1109/access.2021.3107554
fatcat:e3cdraep3rbhbbm76gfpaotbhi
VaryBlock: A Novel Approach for Object Detection in Remote Sensed Images
2019
Sensors
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Object detection, as one of the most challenging tasks in the area of remote sensing, has been remarkably promoted by convolutional neural network (CNN)-based methods like You Only Look Once (YOLO) and ...
We devise and integrate VaryBlock to the architecture which effectively offsets some of the information loss caused by downsampling. ...
In the past few years, convolutional neural network (CNN)-based algorithms have played an important role in AI and achieved flourishing success for computer vision tasks [2] [3] [4] . ...
doi:10.3390/s19235284
pmid:31801269
pmcid:PMC6929156
fatcat:vbwhbcdshrcsvaq3ye2oju4jdm
Recognition and Classification of Broiler Droppings Based on Deep Convolutional Neural Network
2019
Journal of Sensors
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
For comparative purposes, Faster R-CNN and YOLO-V3 deep Convolutional Neural Networks were developed. The performance of YOLO-V3 was improved by optimizing the anchor box. ...
This study proposes an automated broiler digestive disease detector based on a deep Convolutional Neural Network model to classify fine-grained abnormal broiler droppings images as normal and abnormal ...
Anchor Box Selection for YOLO-V3 Based on K-Means+ +.
Conclusion Digestive tract disease is one of the major diseases in broiler breeding. ...
doi:10.1155/2019/3823515
fatcat:uf3nq4x6ubh2dbnrkjkn5bpcci
« Previous
Showing results 1 — 15 out of 3,844 results