• Publications
  • Influence
DecideNet: Counting Varying Density Crowds Through Attention Guided Detection and Density Estimation
TLDR
In real-world crowd counting applications, the crowd densities vary greatly in spatial and temporal domains. Expand
  • 171
  • 16
  • PDF
ARGALI: An Automatic Cup-to-Disc Ratio Measurement System for Glaucoma Analysis Using Level-set Image Processing
TLDR
Glaucoma is a leading cause of blindness worldwide, accounting for 12.3% of the permanently blind according to the World Heath Organization. Expand
  • 76
  • 4
InfAR dataset: Infrared action recognition at different times
TLDR
We construct a new Inf rared A ction R ecognition dataset captured at different times, including in summer and winter, and explore how discriminable actions in our InfAR dataset are with the state-of-the-art pipelines based on low-level features and deep convolutional neural network (CNN), respectively. Expand
  • 28
  • 4
A novel learning-based frame pooling method for event detection
TLDR
We propose a novel learning-based frame pooling method that outperforms the commonly used average pooling and max pooling strategies on both high-level and low-level features. Expand
  • 8
  • 1
  • PDF
A New Dataset and Evaluation for Infrared Action Recognition
TLDR
We construct a new infrared action dataset and evaluate the state-of-the-art AR pipeline, including widely-used low-level local descriptors, coding methods and fusion strategies, on it. Expand
  • 4
  • 1
Small UAV Detection in Videos from a Single Moving Camera
TLDR
We propose a UAV detection method based on a single moving camera to handle the problem for UAVs with fast moving speed. Expand
  • 5
  • 1
People counting based on head detection combining Adaboost and CNN in crowded surveillance environment
TLDR
We propose a head detection based people counting method combining the Adaboost algorithm and the CNN, which can greatly reduce the following classification time. Expand
  • 48
From constrained to unconstrained datasets: an evaluation of local action descriptors and fusion strategies for interaction recognition
TLDR
In this paper, we introduce a new unconstrained dataset, called WEB-interaction, collected from the Internet. Expand
  • 14
Two-Stream Contextualized CNN for Fine-Grained Image Classification
TLDR
We propose a novel approach, two-stream contextualized Convolutional Neural Network, which provides a simple but efficient context-content joint classification model under deep learning framework. Expand
  • 9
  • PDF
1 Gb/s VCSEL/CMOS flip-chip 2-D-array interconnects and associated diffractive optics
980-nm substrate-emitting oxidized-aperture VCSELs with threshold currents in the 150 /spl mu/A to 500 /spl mu/A range have been developed for optoelectronic interconnects. 8/spl times/8 VCSEL arraysExpand
  • 5