Where is my puppy? Retrieving lost dogs by facial features

Moreira, Thierry Pinheiro; Pérez, Mauricio; Werneck, Rafael de Oliveira; Valle, Eduardo

doi:10.1007/s11042-016-3824-1

Cited by 19 publications

(22 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreira et al evaluate the viability of using existing human face recognition methods (EigenFaces, FisherFaces, LBPH, and a Sparse method) as well as deep learning techniques such as Convolutional Neural Networks (BARK and WOOF) for dog recognition [15]. Using a dataset of two different breeds of dogs, huskies and pugs, Moreira et al show that based on using the WOOF model, an accuracy of 75.14% and 54.38% is obtained for huskies and pugs, respectively.…”

Section: Related Workmentioning

confidence: 99%

“…3) Flickr-dog Dataset: Flickr-dog dataset [15] contains two breeds of dogs: pugs and huskies. The dataset contains 42 classes, 21 for each dog breed, up to a total of 374 images with at least 5 images per class.…”

Section: A Datasetmentioning

confidence: 99%

“…We separated the Flickr-dog set into training sets and testing sets following the approach proposed in [15]. We performed 5-fold cross-validation to compute the accuracy of identification.…”

Section: E "Fine Stage": Dog Identificationmentioning

confidence: 99%

See 2 more Smart Citations

Dog Identification using Soft Biometrics and Neural Networks

Lai

Yanushkevich

2019

2019 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

This paper addresses the problem of biometric identification of animals, specifically dogs. We apply advanced machine learning models such as deep neural network on the photographs of pets in order to determine the pet identity. In this paper, we explore the possibility of using different types of "soft" biometrics, such as breed, height, or gender, in fusion with "hard" biometrics such as photographs of the pet's face. We apply the principle of transfer learning on different Convolutional Neural Networks, in order to create a network designed specifically for breed classification. The proposed network is able to achieve an accuracy of 90.80% and 91.29% when differentiating between the two dog breeds, for two different datasets. Without the use of "soft" biometrics, the identification rate of dogs is 78.09% but by using a decision network to incorporate "soft" biometrics, the identification rate can achieve an accuracy of 84.94%.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Dog Identification using Soft Biometrics and Neural Networks

Lai

Yanushkevich

2019

2019 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

show abstract

“…On the other hand, re-identification systems for pets has seen some interest in the computer vision community [3], mainly aiming to retrieving lost "family members".…”

Section: A Animal Re-identification Motivationsmentioning

confidence: 99%

“…Finally, pets represent an opportunity to build larger dataset, as they outnumber the above mentioned animals of a large margin, but public datasets still do not exist, and collecting this data requires huge resources and efforts. As an example, using pictures of two dogs breeds gathered from Flickr, [3] achieved remarkable performances. As dog faces differ from humans ones, the authors developed two Deep CNN trained from scratch on dogs images only, after a pre-processing phase consisting of a tight crop to suppress most of the background.…”

Section: B Animal Re-identificationmentioning

confidence: 99%

Multi-views Embedding for Cattle Re-identification

Bergamini

Porrello

Dondona

et al. 2018

2018 14th International Conference on Signal-Image Technology &Amp; Internet-Based Systems (SITIS)

View full text Add to dashboard Cite

People re-identification task has seen enormous improvements in the latest years, mainly due to the development of better image features extraction from deep Convolutional Neural Networks (CNN) and the availability of large datasets. However, little research has been conducted on animal identification and re-identification, even if this knowledge may be useful in a rich variety of different scenarios. Here, we tackle cattle re-identification exploiting deep CNN and show how this task is poorly related with the human one, presenting unique challenges that makes it far from being solved. We present various baselines, both based on deep architectures or on standard machine learning algorithms, and compared them with our solution. Finally, a rich ablation study has been conducted to further investigate the unique peculiarities of this task.

show abstract

Dog Breed Classification Using CNN

Varghese

Remya

2021

Security Issues and Privacy Concerns in Industry 4.0 Applications

View full text Add to dashboard Cite

Where is my puppy? Retrieving lost dogs by facial features

Cited by 19 publications

References 21 publications

Dog Identification using Soft Biometrics and Neural Networks

Dog Identification using Soft Biometrics and Neural Networks

Multi-views Embedding for Cattle Re-identification

Dog Breed Classification Using CNN

Contact Info

Product

Resources

About