• Publications
  • Influence
Memory Attention Networks for Skeleton-based Action Recognition
TLDR
We propose a temporal-then-spatial recalibration scheme to alleviate such complex variations, resulting in an end-to-end Memory Attention Networks (MANs) which consist of a Temporal Attention Recalibration Module (TARM) and a Spatio-Temporal Convolution Module (STCM). Expand
  • 35
  • 7
  • PDF
A Four-Factor Immunoscore System That Predicts Clinical Outcome for Stage II/III Gastric Cancer
The American Joint Committee on Cancer (AJCC) staging system for gastric cancer is insufficiently prognostic. An immunoscore system based on PD-L1 expression on tumor or immune cells, PD-1Expand
  • 29
  • 2
Deep Fisher discriminant learning for mobile hand gesture recognition
TLDR
Gesture recognition is a challenging problem in the field of biometrics. Expand
  • 33
  • 1
  • PDF
AWBING plus algorithm for generic object proposal generation
  • 1
CMNet: A Connect-and-Merge Convolutional Neural Network for Fast Vehicle Detection in Urban Traffic Surveillance
TLDR
We present a connect-and-merge convolutional neural network (CMNet) for fast detecting vehicles in complex scenes. Expand
  • 6
A quantitative morphometric comparison of cockpit and doline karst landforms
This study presented a quantitative comparison of cockpit and doline karst by examining the numbers and characteristics of typical types of landform entities that are developed in Guilin (Guangxi,Expand
  • 4
Visual trajectory analysis via Replicated Softmax-based models
TLDR
In this paper, we apply the Replicated Softmax model to visual trajectory representation and analysis problems. Expand
  • 1
Gaussian Transfer Convolutional Neural Networks
TLDR
In deep convolutional neural networks (DCNN), the pooling operation is usually adopted to produce condensed and transformation invariant feature maps for the input image. Expand
  • 2
Adaptive bare bones particle swarm optimization for feature selection
TLDR
We propose a new method of feature selection based on an adaptive bare bones particle swarm optimization that achieves better classification accuracy or uses fewer features than the other compared feature selection methods. Expand
  • 2
Few-shot learning with deformable convolution for multi-scale lesion detection in mammography.
TLDR
PURPOSE Image-based breast lesion detection is a powerful clinical diagnosis technology. Expand
  • 1