Smart Mining for Deep Metric Learning

Ben Harwood, B. G. Vijay Kumar, Gustavo Carneiro, Ian Reid, Tom Drummond

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

34 Citations (Scopus)

Abstract

To solve deep metric learning problems and producing feature embeddings, current methodologies will commonly use a triplet model to minimise the relative distance between samples from the same class and maximise the relative distance between samples from different classes. Though successful, the training convergence of this triplet model can be compromised by the fact that the vast majority of the training samples will produce gradients with magnitudes that are close to zero. This issue has motivated the development of methods that explore the global structure of the embedding and other methods that explore hard negative/positive mining. The effectiveness of such mining methods is often associated with intractable computational requirements. In this paper, we propose a novel deep metric learning method that combines the triplet model and the global structure of the embedding space. We rely on a smart mining procedure that produces effective training samples for a low computational cost. In addition, we propose an adaptive controller that automatically adjusts the smart mining hyper-parameters and speeds up the convergence of the training process. We show empirically that our proposed method allows for fast and more accurate training of triplet ConvNets than other competing mining methods. Additionally, we show that our method achieves new state-of-the-art embedding results for CUB-200-2011 and Cars196 datasets.

Original languageEnglish
Title of host publicationProceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017
EditorsRita Cucchiara, Yasuyuki Matsushita, Nicu Sebe, Stefano Soatto
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages2840-2848
Number of pages9
ISBN (Electronic)9781538610329
ISBN (Print)9781538610336
DOIs
Publication statusPublished - 22 Dec 2017
EventIEEE International Conference on Computer Vision 2017 - Venice, Italy
Duration: 22 Oct 201729 Oct 2017
Conference number: 16th
http://iccv2017.thecvf.com/

Conference

ConferenceIEEE International Conference on Computer Vision 2017
Abbreviated titleICCV 2017
CountryItaly
CityVenice
Period22/10/1729/10/17
Internet address

Cite this

Harwood, B., Vijay Kumar, B. G., Carneiro, G., Reid, I., & Drummond, T. (2017). Smart Mining for Deep Metric Learning. In R. Cucchiara, Y. Matsushita, N. Sebe, & S. Soatto (Eds.), Proceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017 (pp. 2840-2848). Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICCV.2017.307
Harwood, Ben ; Vijay Kumar, B. G. ; Carneiro, Gustavo ; Reid, Ian ; Drummond, Tom. / Smart Mining for Deep Metric Learning. Proceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017. editor / Rita Cucchiara ; Yasuyuki Matsushita ; Nicu Sebe ; Stefano Soatto. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2017. pp. 2840-2848
@inproceedings{c9b9666fc37840ec832721ac4a1b5d3a,
title = "Smart Mining for Deep Metric Learning",
abstract = "To solve deep metric learning problems and producing feature embeddings, current methodologies will commonly use a triplet model to minimise the relative distance between samples from the same class and maximise the relative distance between samples from different classes. Though successful, the training convergence of this triplet model can be compromised by the fact that the vast majority of the training samples will produce gradients with magnitudes that are close to zero. This issue has motivated the development of methods that explore the global structure of the embedding and other methods that explore hard negative/positive mining. The effectiveness of such mining methods is often associated with intractable computational requirements. In this paper, we propose a novel deep metric learning method that combines the triplet model and the global structure of the embedding space. We rely on a smart mining procedure that produces effective training samples for a low computational cost. In addition, we propose an adaptive controller that automatically adjusts the smart mining hyper-parameters and speeds up the convergence of the training process. We show empirically that our proposed method allows for fast and more accurate training of triplet ConvNets than other competing mining methods. Additionally, we show that our method achieves new state-of-the-art embedding results for CUB-200-2011 and Cars196 datasets.",
author = "Ben Harwood and {Vijay Kumar}, {B. G.} and Gustavo Carneiro and Ian Reid and Tom Drummond",
year = "2017",
month = "12",
day = "22",
doi = "10.1109/ICCV.2017.307",
language = "English",
isbn = "9781538610336",
pages = "2840--2848",
editor = "Rita Cucchiara and Yasuyuki Matsushita and Nicu Sebe and Stefano Soatto",
booktitle = "Proceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States of America",

}

Harwood, B, Vijay Kumar, BG, Carneiro, G, Reid, I & Drummond, T 2017, Smart Mining for Deep Metric Learning. in R Cucchiara, Y Matsushita, N Sebe & S Soatto (eds), Proceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017. IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 2840-2848, IEEE International Conference on Computer Vision 2017, Venice, Italy, 22/10/17. https://doi.org/10.1109/ICCV.2017.307

Smart Mining for Deep Metric Learning. / Harwood, Ben; Vijay Kumar, B. G.; Carneiro, Gustavo; Reid, Ian; Drummond, Tom.

Proceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017. ed. / Rita Cucchiara; Yasuyuki Matsushita; Nicu Sebe; Stefano Soatto. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 2840-2848.

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

TY - GEN

T1 - Smart Mining for Deep Metric Learning

AU - Harwood, Ben

AU - Vijay Kumar, B. G.

AU - Carneiro, Gustavo

AU - Reid, Ian

AU - Drummond, Tom

PY - 2017/12/22

Y1 - 2017/12/22

N2 - To solve deep metric learning problems and producing feature embeddings, current methodologies will commonly use a triplet model to minimise the relative distance between samples from the same class and maximise the relative distance between samples from different classes. Though successful, the training convergence of this triplet model can be compromised by the fact that the vast majority of the training samples will produce gradients with magnitudes that are close to zero. This issue has motivated the development of methods that explore the global structure of the embedding and other methods that explore hard negative/positive mining. The effectiveness of such mining methods is often associated with intractable computational requirements. In this paper, we propose a novel deep metric learning method that combines the triplet model and the global structure of the embedding space. We rely on a smart mining procedure that produces effective training samples for a low computational cost. In addition, we propose an adaptive controller that automatically adjusts the smart mining hyper-parameters and speeds up the convergence of the training process. We show empirically that our proposed method allows for fast and more accurate training of triplet ConvNets than other competing mining methods. Additionally, we show that our method achieves new state-of-the-art embedding results for CUB-200-2011 and Cars196 datasets.

AB - To solve deep metric learning problems and producing feature embeddings, current methodologies will commonly use a triplet model to minimise the relative distance between samples from the same class and maximise the relative distance between samples from different classes. Though successful, the training convergence of this triplet model can be compromised by the fact that the vast majority of the training samples will produce gradients with magnitudes that are close to zero. This issue has motivated the development of methods that explore the global structure of the embedding and other methods that explore hard negative/positive mining. The effectiveness of such mining methods is often associated with intractable computational requirements. In this paper, we propose a novel deep metric learning method that combines the triplet model and the global structure of the embedding space. We rely on a smart mining procedure that produces effective training samples for a low computational cost. In addition, we propose an adaptive controller that automatically adjusts the smart mining hyper-parameters and speeds up the convergence of the training process. We show empirically that our proposed method allows for fast and more accurate training of triplet ConvNets than other competing mining methods. Additionally, we show that our method achieves new state-of-the-art embedding results for CUB-200-2011 and Cars196 datasets.

UR - http://www.scopus.com/inward/record.url?scp=85041913153&partnerID=8YFLogxK

U2 - 10.1109/ICCV.2017.307

DO - 10.1109/ICCV.2017.307

M3 - Conference Paper

AN - SCOPUS:85041913153

SN - 9781538610336

SP - 2840

EP - 2848

BT - Proceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017

A2 - Cucchiara, Rita

A2 - Matsushita, Yasuyuki

A2 - Sebe, Nicu

A2 - Soatto, Stefano

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

ER -

Harwood B, Vijay Kumar BG, Carneiro G, Reid I, Drummond T. Smart Mining for Deep Metric Learning. In Cucchiara R, Matsushita Y, Sebe N, Soatto S, editors, Proceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2017. p. 2840-2848 https://doi.org/10.1109/ICCV.2017.307