Learning hierarchical similarity metrics

Verma, Nakul; Mahajan, Dhruv; Sellamanickam, Sundararajan; Nair, Varun Sasidharan

doi:10.1109/cvpr.2012.6247938

Cited by 98 publications

(86 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[42], Verma et al . [12]. However, they assume the presence of a taxonomy (most often a natural semantic taxonomy), while here we do not assume any such information.…”

Section: Context and Related Workmentioning

confidence: 99%

“…Interestingly, they show that nearest neighbor classifiers using the learned metrics get improved performance over Euclidean distancebased k-NN and over discriminative methods. Our approach bears similarity with [12] as we also learn a hierarchy of similarity metrics. However, a notable difference is that our approach does not require any taxonomy.…”

Section: : Input: (I) Set Of Face Featuresmentioning

confidence: 99%

“…[12], who proposed a framework for learning hierarchical similarity metrics using class taxonomies. Interestingly, they show that nearest neighbor classifiers using the learned metrics get improved performance over Euclidean distancebased k-NN and over discriminative methods.…”

Section: : Input: (I) Set Of Face Featuresmentioning

confidence: 99%

“…1 gives an overview of our approach in contrast to traditional metric learning. One big difference with [10,11] or [12] is that our approach does not need any face taxonomy nor predefined set of attributes. Both are automatically discovered.…”

Section: Introductionmentioning

confidence: 99%

“…In the same way, Verma et al . [12] proposed a novel framework to learn similarity metrics using class taxonomies, showing that nearest neighbor classifiers using the learned metrics get improved performance over the best discriminative methods. Inspired by these previous works, we propose to organize large face databases hierarchically using locally and discriminatively learned projections.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Some Faces are More Equal than Others: Hierarchical Organization for Accurate and Efficient Large-Scale Identity-Based Face Retrieval

Bhattarai

Sharma

Jurie

et al. 2015

Computer Vision - ECCV 2014 Workshops

View full text Add to dashboard Cite

Abstract. This paper presents a novel method for hierarchically organizing large face databases, with application to efficient identity-based face retrieval. The method relies on metric learning with local binary pattern (LBP) features. On one hand, LBP features have proved to be highly resilient to various appearance changes due to illumination and contrast variations while being extremely efficient to calculate. On the other hand, metric learning (ML) approaches have been proved very successful for face verification 'in the wild', i.e. in uncontrolled face images with large amounts of variations in pose, expression, appearances, lighting, etc. While such ML based approaches compress high dimensional features into low dimensional spaces using discriminatively learned projections, the complexity of retrieval is still significant for large scale databases (with millions of faces). The present paper shows that learning such discriminative projections locally while organizing the database hierarchically leads to a more accurate and efficient system. The proposed method is validated on the standard Labeled Faces in the Wild (LFW) benchmark dataset with millions of additional distracting face images collected from photos on the internet.

show abstract

“…[42], Verma et al . [12]. However, they assume the presence of a taxonomy (most often a natural semantic taxonomy), while here we do not assume any such information.…”

Section: Context and Related Workmentioning

confidence: 99%

Section: : Input: (I) Set Of Face Featuresmentioning

confidence: 99%

Section: : Input: (I) Set Of Face Featuresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Some Faces are More Equal than Others: Hierarchical Organization for Accurate and Efficient Large-Scale Identity-Based Face Retrieval

Bhattarai

Sharma

Jurie

et al. 2015

Computer Vision - ECCV 2014 Workshops

View full text Add to dashboard Cite

show abstract

Contemplating Visual Emotions: Understanding and Overcoming Dataset Bias

Panda

Zhang

Li³

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

While machine learning approaches to visual emotion recognition offer great promise, current methods consider training and testing models on small scale datasets covering limited visual emotion concepts. Our analysis identifies an important but long overlooked issue of existing visual emotion benchmarks in the form of dataset biases. We design a series of tests to show and measure how such dataset biases obstruct learning a generalizable emotion recognition model. Based on our analysis, we propose a webly supervised approach by leveraging a large quantity of stock image data. Our approach uses a simple yet effective curriculum guided training strategy for learning discriminative emotion features. We discover that the models learned using our large scale stock image dataset exhibit significantly better generalization ability than the existing datasets without the manual collection of even a single label. Moreover, visual representation learned using our approach holds a lot of promise across a variety of tasks on different image and video datasets. 4 The image is taken from Google Images with the search keyword sad amusement park. Source: https://goo.gl/AUwoPZ

show abstract