• Publications
  • Influence
NUS-WIDE: a real-world web image database from National University of Singapore
TLDR
This paper introduces a web image dataset created by NUS's Lab for Media Search, with 269,648 images and the associated tags from Flickr, with a total of 5,018 unique tags. Expand
  • 1,820
  • 528
  • PDF
Nonconvex Nonsmooth Low Rank Minimization via Iteratively Reweighted Nuclear Norm
TLDR
We propose to use a family of nonconvex surrogates of L0-norm on the singular values of a matrix to approximate the rank function. Expand
  • 164
  • 27
  • PDF
Generalized Nonconvex Nonsmooth Low-Rank Minimization
TLDR
We propose an Iteratively Reweighted Nuclear Norm (IRNN) algorithm to solve the nonconvex nonsmooth low-rank minimization problem. Expand
  • 197
  • 24
  • PDF
Outer Product-based Neural Collaborative Filtering
TLDR
We propose to use an outer product to explicitly model the pairwise correlations between embedding dimensions of the embedding space in a multi-layer neural recommender model. Expand
  • 132
  • 19
  • PDF
Richer Convolutional Features for Edge Detection
TLDR
We propose richer convolutional features (RCF), a novel deep structure fully exploiting the CNN features from all the conv layers, to perform the pixel-wise prediction for edge detection in an image-to-image fashion. Expand
  • 187
  • 18
  • PDF
Camera Constraint-Free View-Based 3-D Object Retrieval
TLDR
A camera constraint-free view-based (CCFV) 3-D object retrieval algorithm is proposed in this paper. Expand
  • 213
  • 18
  • PDF
Label to region by bi-layer sparsity priors
TLDR
We propose a bi-layer sparse coding formulation for uncovering how an image or semantic region can be robustly reconstructed from the over-segmented image patches of an image set. Expand
  • 89
  • 18
  • PDF
Unified Video Annotation via Multigraph Learning
TLDR
We propose a method named optimized multigraph-based semi-supervised learning (OMG-SSL), which aims to simultaneously tackle these difficulties in a unified scheme. Expand
  • 446
  • 13
  • PDF
Correlative multi-label video annotation
TLDR
We propose a third paradigm which simultaneously classifies concepts and models correlations between them in a single step by using a novel Correlative Multi-Label (CML) framework. Expand
  • 453
  • 13
  • PDF
Weakly-Supervised Dual Clustering for Image Semantic Segmentation
TLDR
We propose a novel Weakly-Supervised Dual Clustering approach for image semantic segmentation with image-level labels, i.e., collaboratively performing image segmentation and tag alignment with those regions. Expand
  • 73
  • 13
  • PDF