• Publications
  • Influence
Discrimination-aware Channel Pruning for Deep Neural Networks
TLDR
We investigate a simple-yet-effective method, called discrimination-aware channel pruning, to choose those channels that really contribute to discriminative power. Expand
  • 184
  • 22
  • PDF
Towards Context-Aware Interaction Recognition for Visual Relationship Detection
TLDR
This paper proposes an alternative, context-aware interaction recognition framework which combines the context, and the interaction. Expand
  • 72
  • 15
  • PDF
Towards Effective Low-Bitwidth Convolutional Neural Networks
TLDR
This paper tackles the problem of training a deep convolutional neural network with both low-precision weights and low-bitwidth activations. Expand
  • 80
  • 8
  • PDF
Attend in Groups: A Weakly-Supervised Deep Learning Framework for Learning from Web Data
TLDR
We propose an end-to-end weakly-supervised deep learning framework to learn visual representions from massive Web data without any human supervision. Expand
  • 56
  • 8
  • PDF
Sequential Person Recognition in Photo Albums with a Recurrent Network
TLDR
Recognizing the identities of people in everyday photos is still a very challenging problem for machine vision, due to issues such as non-frontal faces, changes in clothing, location, lighting. Expand
  • 25
  • 8
  • PDF
TasselNet: counting maize tassels in the wild via local counts regression network
TLDR
TasselNet can achieve robust in-field counting of maize tassels with a relatively high degree of accuracy and outperforms other state-of-the-art approaches. Expand
  • 60
  • 7
  • PDF
Structured Binary Neural Networks for Accurate Image Classification and Semantic Segmentation
TLDR
We propose to train convolutional neural networks with both binarized weights and activations, leading to quantized models specifically for mobile devices with limited power capacity and computation resources. Expand
  • 47
  • 7
  • PDF
HCVRD: A Benchmark for Large-Scale Human-Centered Visual Relationship Detection
TLDR
We propose a large-scale human-centric visual relationship detection dataset, which provides many more types of relationship annotations (nearly 10K categories) than the previous released datasets. Expand
  • 13
  • 6
Visual Tracking via Discriminative Sparse Similarity Map
TLDR
In this paper, we cast the tracking problem as finding the candidate that scores highest in the evaluation model based upon a matrix called discriminative sparse similarity map which is obtained via a multi-task reverse sparse coding approach with Laplacian constraint. Expand
  • 74
  • 5
Fast Training of Triplet-Based Deep Binary Embedding Networks
TLDR
In this paper, we aim to learn a mapping (or embedding) from images to a compact binary space in which Hamming distances correspond to a ranking measure for image retrieval task. Expand
  • 73
  • 5
  • PDF