• Publications
  • Influence
Learning Social Tag Relevance by Neighbor Voting
TLDR
Social image analysis and retrieval is important for helping people organize and access the increasing amount of user tagged multimedia. Expand
  • 373
  • 52
  • PDF
Deep Text Classification Can be Fooled
TLDR
In this paper, we present an effective method to craft text adversarial samples, revealing one important yet underestimated fact that DNN-based text classifiers are also prone to adversarial sample attack. Expand
  • 167
  • 27
  • PDF
Annotating Images by Mining Image Search Results
TLDR
In this paper, we propose a novel attempt at model-free image annotation, which is a data-driven approach that annotates images by mining their search results. Expand
  • 199
  • 14
  • PDF
Detecting Adversarial Image Examples in Deep Neural Networks with Adaptive Noise Reduction
TLDR
We propose a straightforward method for detecting adversarial image examples, which can be directly deployed into unmodified off-the-shelf DNN models. Expand
  • 84
  • 13
  • PDF
Learning tag relevance by neighbor voting for social image retrieval
TLDR
We propose a novel algorithm that scalably and reliably learns tag relevance by accumulating votes from visually similar neighbors. Expand
  • 194
  • 11
  • PDF
The MediaMill TRECVID 2006 Semantic Video Search Engine
TLDR
In this paper we describe our TRECVID 2008 video retrieval experiments. Expand
  • 322
  • 9
  • PDF
Word2VisualVec: Image and Video to Sentence Matching by Visual Feature Prediction
TLDR
We propose Word2VisualVec, a deep neural network architecture that learns to predict a deep visual encoding of textual input based on sentence vectorization and a multi-layer perceptron. Expand
  • 34
  • 8
  • PDF
Personalizing automated image annotation using cross-entropy
TLDR
We propose a cross-entropy based learning algorithm which personalizes a generic annotation model by learning from a user's multimedia tagging history. Expand
  • 32
  • 8
  • PDF
Socializing the Semantic Gap
TLDR
We survey the state of the art of content- based image retrieval in the context of social image platforms and tagging, with a comprehensive treatise of the closely linked problems of image tag assignment, image tag refinement, and tag-based image retrieval. Expand
  • 145
  • 7
  • PDF
Image annotation by large-scale content-based image retrieval
TLDR
In this paper, we target at solving the automatic image annotation problem in a novel search and mining framework that efficiently leverages large-scale and well-annotated images, and is potentially capable of dealing with unlimited vocabulary. Expand
  • 122
  • 7
  • PDF