Research Behavior Analysis & Understanding
Gaze Estimation by Tracking Iris and Eyelid using a Monocular Camera
Human’s gaze information includes user’s idea and interest, so it is useful for drivers’ active safety system, marketing tools for digital signage and so on. In these situations, it is needed that everyone can use it instantly. So, there is a great demand for an easy and useful Gaze Estimation system.

Research contents
-
Person Re-identification using Distance Learning by Convolutional Neural Networks
We propose a distance learning method using Convolutional Neural Networks to re-identify people based on moving image data obtained...
-
Multiple Object Extraction from Aerial Imagery with Convolutional Neural Networks
An automatic system to extract terrestrial objects from aerial imagery has many applications in a wide range of areas. ...
-
People Tracking in Dense Crowds using Optical Flow Clustering based Superpixel
People tracking in crowded scenes should be focused since the problem is beneficial and challenging in computer vision field. ...
-
Moving Object Proposal by Grouping with Motion Feature
People tracking in crowded scenes should be focused since the problem is beneficial and challenging in computer vision field. However, the problem of “crowd tracking” is extremely difficult ...
-
Dilated Convolutions for Image Classification and Object Localization
Dilated convolutions are very effective in dense prediction problems such as semantic segmentation. ...
Behavior Analysis & Understanding
-
Top-View Human Pose Estimation via Convolutional Neural Networks
We propose the DeepPose-based pose estimation system that is flexible with the change of bounding-box range for top-view images. ...
-
Detecting Conversational Group Associates from Static
We propose a novel method for estimating levels of conversational involvement. With the recent development of single-person analysis in computer vision,
-
Action detection of TV viewer by object information and image feature
In this research, we propose a method to detect ながら見行動 which is thought to be important in estimating the degree of concentration on contents when watching TV.
-
In this research, we propose a behavior recognition method that is ...
-
Gaze Estimation by Tracking Iris and Eyelid using a Monocular Camera
Human“s gaze information includes user“s idea and interest, so it is useful for drivers“ active safety system,
-
In the proposed method, we first estimate the location of the human body part of the driver using depth learning using two modals, an IR image and a depth image as input.
-
Swimmer detection and stroke rate estimation using swimmer’s motion intensity
The analysis of sports videos has been a heavily investigated subject in computer vision. Among sports videos,...
-
Automatic Match Analysis in Sports Videos Using Bayesian Network and Random Forest
There are many automatic match analysis methods that use video information. Our goal is to improve the precision of previous match analysis methods. ...
-
Form analysis by unsupervised classification of tennis videos using posture estimation
In this research, we propose a method to classify pose information obtained by estimating joint position of players from tennis game images by classification by unsupervised classification method, ...
-
We propose a per-frame upper body pose estimation method...
-
Play analysis for rugby video by using time series Deep Learning and multimodal information
For the purpose of constructing a sports image analysis system using image recognition technology, ...
-
6D Object Pose Estimation using 3D Vector Pairs
This method is a stable 3D object detection that can be applied to complicated scenes consisting of randomly stacked industrial parts. ...
-
Physical Reasoning for 3D Object Recognition using Global Hypothesis Verification
We have proposed a method to recognize the 6DoF pose of multiple objects simultaneously. ...
-
The mechanism behind collisions between vehicles and pedestrians must be thoroughly studied in order to prevent future traffic accidents. ...
-
Pedestrian Near-Miss Analysis on Vehicle-Mounted Driving Recorders
Recently, a demand for video analysis on vehicle- mounted driving recorders has been increasing in vision-based safety systems, such as for autonomous vehicles. ...
-
The CNN Based Spine and Cobb Angle Estimator from Moire Photography
Adolescent idiopathic scoliosis (AIS) causes serious health problems when left untreated after accruing. In Japan, ...
-
Hand-held HDR with Removing Blur Due to Camera Shake
We proposed a HDR image reconstruction method from bracket captured inputs with hand-held Camera. ...
-
Depth Image Enhancement Using Local Tangent Plane Approximations
We propose a depth image enhancement method for consumer RGB-D cameras. Most existing methods use the pixel-coordinates of the aligned color image. ...
-
Change detection via semantic segmentation and difference extraction method
We aim to develop a fast and accurate method of updating maps by detecting local differences occurred over different time series; ...