• Publications
  • Influence
Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?
TLDR
The purpose of this study is to determine whether current video datasets have sufficient data for training very deep convolutional neural networks (CNNs) with spatio-temporal three-dimensional (3D) kernels. Expand
  • 634
  • 139
  • PDF
Learning Spatio-Temporal Features with 3D Residual Networks for Action Recognition
TLDR
We explore the effectiveness of ResNets with 3D convolutional kernels for action recognition. Expand
  • 170
  • 24
  • PDF
Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB
TLDR
We propose a novel approach for traffic accident anticipation through (i) Adaptive Loss for Early Anticipation (AdaLEA) and (ii) a large-scale self-annotated incident database for anticipation. Expand
  • 34
  • 9
  • PDF
Robust object detection using a Radial Reach Filter (RRF)
TLDR
In this paper, the authors propose a new filter process called a Radial Reach Filter (RRF) in order to extract reliably an object from its background using a background image. Expand
  • 14
  • 6
Robust event detection by radial reach filter (RRF)
TLDR
We propose a novel statistical measure for robust event detection, called 'Radial Reach filter' (RRF). Expand
  • 26
  • 5
Using selective correlation coefficient for robust image registration
TLDR
A new method is proposed for robust image registration named Selective Correlation Coefficient in order to search images under ill-conditioned illumination or partial occlusion. Expand
  • 116
  • 4
Robust object detection using a Radial Reach Filter (RRF)
TLDR
In this paper the authors report on a new algorithm used to separate an object from its background using a background image using a Radial Reach Filter (RRF). Expand
  • 19
  • 3
CT findings of leukemic pulmonary infiltration with pathologic correlation
The aim of this study was to demonstrate the characteristic CT findings of leukemic pulmonary infiltration based on the pathologic findings. The CT findings of 11 leukemic patients with leukemicExpand
  • 50
  • 2
Robust object detection and segmentation by peripheral increment sign correlation image
TLDR
This paper proposes the peripheral increment sign correlation image, which can be used to evaluate sign changes of the neighborhood brightness. Expand
  • 18
  • 2
Human Action Recognition Without Human
TLDR
The objective of this paper is to evaluate “human action recognition without human”. Expand
  • 20
  • 2
  • PDF