ImageCLEF 2014: Overview and Analysis of the Results

Caputo, Barbara; Müller, Henning; Martínez‐Gómez, Jesús; Villegas, Mauricio; Açar, Burak; Patricia, Novi; Marvasti, Neda Barzegar; Uskudarli, Suzan; Paredes, Roberto; Cazorla, Miguel; García-Varea, Ismael; Morell, Vicente

doi:10.1007/978-3-319-11382-1_18

Cited by 64 publications

(30 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ImageCLEF-DA [4] originally used for the ImageCLEF 2014 domain adaptation challenge consists of twelve common classes from three domains: ImageNet ILSVRC 2012 (I), Pascal VOC 2012 (P), and Caltech-256 (C). Each doamin has 600 images in total and contains 50 images per class.…”

Section: Datasets and Baselinesmentioning

confidence: 99%

Progressive Feature Alignment for Unsupervised Domain Adaptation

Chen

Xie

Huang

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

394

239

View full text Add to dashboard Cite

Unsupervised domain adaptation (UDA) transfers knowledge from a label-rich source domain to a fullyunlabeled target domain. To tackle this task, recent approaches resort to discriminative domain transfer in virtue of pseudo-labels to enforce the class-level distribution alignment across the source and target domains. These methods, however, are vulnerable to the error accumulation and thus incapable of preserving cross-domain category consistency, as the pseudo-labeling accuracy is not guaranteed explicitly. In this paper, we propose the Progressive Feature Alignment Network (PFAN) to align the discriminative features across domains progressively and effectively, via exploiting the intra-class variation in the target domain. To be specific, we first develop an Easyto-Hard Transfer Strategy (EHTS) and an Adaptive Prototype Alignment (APA) step to train our model iteratively and alternatively. Moreover, upon observing that a good domain adaptation usually requires a non-saturated source classifier, we consider a simple yet efficient way to retard the convergence speed of the source classification loss by further involving a temperature variate into the soft-max function. The extensive experimental results reveal that the proposed PFAN exceeds the state-of-the-art performance on three UDA datasets. * indicates equal contributions.† Corresponding authors

show abstract

Section: Datasets and Baselinesmentioning

confidence: 99%

Progressive Feature Alignment for Unsupervised Domain Adaptation

Chen

Xie

Huang

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

394

239

View full text Add to dashboard Cite

show abstract

“…We conducted our experiment on the WEBUPV dataset of Scalable Concept Image Annotation subtask of the Image-CLEF 2014 [19]. The dataset contains three sets of image data; the training set, the development set, and the test set.…”

Section: A Datasetmentioning

confidence: 99%

“…However, most of them concentrate on scalability issue; while accuracy is still poor, semantic gap is still remain, and by some means depend on expensive human-labeled data. Another effort to reduce the dependencies of human-labeled image data has been taken by ImageCLEF [19] forum. This forum has been organizing the photo annotation and retrieval task [20], [21] for the last several years, where training data is a large collection of Web images without ground truth labels.…”

Section: Related Workmentioning

confidence: 99%

Ontology based classification for multi-label image annotation

Reshma

Ullah

Aono

2014

2014 International Conference of Advanced Informatics: Concept, Theory and Application (ICAICTA)

View full text Add to dashboard Cite

Image annotation has been an important task for visual information retrieval. It usually involves a multi-class multi-label classification problem. To solve this problem, many researches have been conducted during last two decades, although most of the proposed methods rely on the training data with the ground truth. To prepare such a ground truth is an expensive and laborious task that cannot be easily scaled, and "semantic gaps" between low-level visual features and high-level semantics still remain. In this paper, we propose a novel approach, ontology based supervised learning for multi-label image annotation, where classifiers' training is conducted using easily gathered Web data. Moreover, it takes advantage of both low-level visual features and high-level semantic information of given images. Experimental results using 0.507 million Web images database show effectiveness of the proposed framework over existing method.

show abstract

“…Each RGB-D image is annotated with the semantic category of the room it was acquired, from a set of ten room categories. Unreleased sequences from ViDRILO have been successfully used in the RobotVision at Image-CLEF competition [13] in 2013 [3] and 2014 [2]. Fig.…”

Section: Dataset Description: Vidrilomentioning

confidence: 99%

Semantic localization in the PCL library

Martínez‐Gómez

Morell

Cazorla

et al. 2016

Robotics and Autonomous Systems

Self Cite

View full text Add to dashboard Cite

The semantic localization problem in robotics consists in determining the place where a robot is located by means of semantic categories. The problem is usually addressed as a supervised classification process, where input data correspond to robot perceptions while classes to semantic categories, like kitchen or corridor.In this paper we propose a framework, implemented in the PCL library, which provides a set of valuable tools to easily develop and evaluate semantic localization systems. The implementation includes the generation of 3D global descriptors following a Bag-of-Words approach. This allows the generation of dimensionality-fixed descriptors from any type of keypoint detector and feature extractor combinations. The framework has been designed, structured and implemented in order to be easily extended with different keypoint detectors, feature extractors as well as classification models.The proposed framework has also been used to evaluate the performance of a set of already implemented descriptors, when used as input for a specific semantic localization system. The results obtained are discussed paying special attention to the internal parameters of the BoW descriptor generation process. Moreover, we also review the combination of some keypoint detectors with different 3D descriptor generation techniques.

show abstract

ImageCLEF 2014: Overview and Analysis of the Results

Cited by 64 publications

References 25 publications

Progressive Feature Alignment for Unsupervised Domain Adaptation

Progressive Feature Alignment for Unsupervised Domain Adaptation

Ontology based classification for multi-label image annotation

Semantic localization in the PCL library

Contact Info

Product

Resources

About