Image Clustering Using Local Discriminant Models and Global Integration

By Yi Yang, Dong Xu, Feiping Nie, Shuicheng Yan, Yueting Zhuang. 2010

402 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Unsupervised Deep Embedding for Clustering Analysis (2015, methodology): “We evaluate the proposed method (DEC) on one text dataset and two image datasets and compare it against other algorithms including k-means, LDGMI (Yang et al., 2010), and SEC (Nie et al.”
Soft Filter Pruning for Accelerating Deep Convolutional Neural Networks (2018, neutral): “SFP should also be effective on different computer vision tasks, such as [ Kang et al. , 2017; Ren et al. , 2015; Dong et al. , 2018; Shen et al. , 2018b; Yang et al. , 2010; Shen et al. , 2018a; Dong et al. , 2017b ] , and we will explore this in future.”
Feature Selection (2016, methodology): “Following the definition of global discriminative information (Yang et al. 2010; Fukunaga 2013), the local discriminative score for each instance xi is DSi = tr [(S (i ) t + λId ) −1S(i ) b ] = tr [W′X′P(i )X̃i ′ (X̃i X̃i ′ + λId ) −1X̃i P ′ (i )XW].”
Deep Adaptive Image Clustering (2017, background): “In the literature, much research has been dedicated to image clustering [9, 29, 32, 37, 38].”
Knowledge-based Systems (2009, neutral): “The alternatives over which participants make value-laden choices are the policy options presented in the PVE (Table 1): 𝑂 = { 𝑜 Considering the four aspects, we define a value system in a PVE as follows.”
Structural Deep Clustering Network (2020, background): “into the same category [5, 21]. Over the past decades, a large family of clustering algorithms has been developed and successfully applied to various real-world applications, such as image clustering [28] and text clustering [1]. Recently, the breakthroughs in deep learning have led to a paradigm sh”
A General and Adaptive Robust Loss Function (2017, neutral): “L D M G I[ 36 ]”
Spectral Embedded Clustering: A Framework for In-Sample and Out-of-Sample Spectral Clustering (2011, neutral): “Moreover, this Laplacian matrix Ll is similar to that in [35] which is based on local discriminative analysis, however, Ll is derived from local regression.”
Deep Clustering via Joint Convolutional Autoencoder Embedding and Relative Entropy Minimization (2017, methodology): “For example, DEC , SEC , and LDMGI report their best results by tuning one hyper-parameter over nine different options, and JULE-SF and JULE-RC achieve their good performance by tweak-ing several hyper-parameters over various datasets.”
Multi-class Classification without Multi-class Labels (2019, methodology): “…is by using a part of the training data with class labels and assigning the output nodes to the dominant class which activates the node (here we obtain the optimal assignment by the Hungarian algorithm (Kuhn, 1955), which is commonly used in evaluating the clustering accuracy (Yang et al., 2010)).”
A Multimedia Retrieval Framework Based on Semi-Supervised Ranking and Relevance Feedback (2012, methodology): “First, the Laplacian matrix proposed in this paper can be also interpreted in terms of local discriminative analysis [33], and potentially applied to many other applications, such as clustering [33], semi-supervised classification, etc.”
Clustering-Guided Sparse Structural Learning for Unsupervised Feature Selection (2014, methodology): “, a subset of MNIST used in [43], Binary Alphabet (BA) [40] and a subset of USPS with 40 samples randomly selected for each class [40]), three text data sets (i.”
Unifying Virtual and Physical Worlds (2017, background): “The early fusion methods concatenate all views into a single view and then adapt any conventional off-the-shelf clustering algorithms, such as k-means [Blaschko and Lampert 2008; Chaudhuri et al. 2009].”
Deep Image Category Discovery using a Transferred Similarity Function (2016, background): “The ﬁrst is unsupervised clustering accuracy (ACC) [28] and the second is normalized-mutual information (NMI) [24].”
Graph Learning for Multiview Clustering (2018, neutral): “Besides with Gaussian function, there are several graph construction methods, such as local linear similarity graph [1], k-nearest neighbor graph [2], [3], local discriminant graph [4], [5], pairwise similarity graph [6], and graph learned with subspace clustering [7].”
Variational Deep Embedding: An Unsupervised and Generative Approach to Clustering (2016, methodology): “We also show the performance of using GMM directly on the observed space (GMM), using VAE on the observed space and then using GMM on the latent space from VAE (VAE+GMM)4, as well as the performances of LDMGI [Yang et al., 2010], AAE [Makhzani et al.”
Robust continuous clustering (2017, background): “For reference, we also show the corresponding timings for afﬁnity propagation, a well-known modern clustering algorithm (10), and LDMGI, the baseline that demonstrated the best performance across datasets (48).”