Empirical investigation of active learning strategies
Many predictive tasks require labeled data to induce classification models. The data labeling
process may have a high cost. Several strategies have been proposed to optimize the …
process may have a high cost. Several strategies have been proposed to optimize the …
[PDF][PDF] Semi-supervised active learning with cross-class sample transfer.
To save the labeling efforts for training a classification model, we can simultaneously adopt
Active Learning (AL) to select the most informative samples for human labeling, and Semi …
Active Learning (AL) to select the most informative samples for human labeling, and Semi …
Fast view-based 3D model retrieval via unsupervised multiple feature fusion and online projection learning
Since each visual feature only reflects a unique characteristic about a 3-dimensional (3D)
model and different visual features have diverse discriminative power in model …
model and different visual features have diverse discriminative power in model …
Manifold optimal experimental design via dependence maximization for active learning
Naturally occurring data have been growing in a huge volume size, which poses a big
challenge to give them high-quality labels to learn a good model. Therefore, it is critical to …
challenge to give them high-quality labels to learn a good model. Therefore, it is critical to …
[PDF][PDF] Image Tag Recommendation Algorithm Using Tensor Factorization.
W Yong-sheng - Journal of Multimedia, 2014 - Citeseer
This paper aims to provide high quality tags for digital images according to users' interest. As
there are three main elements in image tag recommendation problem, tensor factorization …
there are three main elements in image tag recommendation problem, tensor factorization …
Unsupervised dual learning for feature and instance selection
Feature selection and instance selection are dual operations on a data matrix. Feature
selection aims at selecting a subset of relevant and informative features from original feature …
selection aims at selecting a subset of relevant and informative features from original feature …
Color Features and Color Spaces Applications to the Automatic Image Annotation
V Maihami, F Yaghmaee - … in Intelligent Applications for Image and …, 2016 - igi-global.com
Nowadays images play a crucial role in different fields such as medicine, advertisement,
education and entertainment. Describing images content and retrieving them are important …
education and entertainment. Describing images content and retrieving them are important …
[PDF][PDF] Using rough set theory to improve content based image retrieval system
MS Lotfabadi, MF Shiratuddin, KW Wong - 2016 - researchportal.murdoch.edu.au
Each image in a Content Based Image Retrieval (CBIR) system is represented by its
features such as colour, texture and shape. These three groups of features are stored in the …
features such as colour, texture and shape. These three groups of features are stored in the …
Cross-Media Feature Learning Framework with Semi-supervised Graph Regularization
T Qi, H Zhang, G Dai - … in Multimedia Information Processing–PCM 2018 …, 2018 - Springer
With the development of multimedia data, cross-media retrieval has become increasingly
important. It can provide the retrieval results with various types of media at the same time by …
important. It can provide the retrieval results with various types of media at the same time by …
[PDF][PDF] Systèmes interactifs par apprentissage actif pour la reconnaissance de gestes
T Sanchez - 2018 - atiam.ircam.fr
Nous observons que la stratégie d'apprentissage non-agnostique basée sur l'incertitude du
modèle permet un apprentissage plus efficace dans les premières requêtes qu'une stratégie …
modèle permet un apprentissage plus efficace dans les premières requêtes qu'une stratégie …