Naohiko Suzuki

Learn More
We propose a method for real-time people tracking using multiple cameras. The particle filter framework is known to be effective for tracking people, but most of existing methods adopt only simple perceptual cues such as color histogram or contour similarity for hypothesis evaluation. To improve the robustness and accuracy of tracking more sophisticated(More)
あらまし 視野を共有する複数のカメラを用いて,三次元空間における人物の実時間追跡を行う.従来より, パーティクルフィルタを用いた人物追跡手法の有効性が報告されている.しかし,観測による仮説の評価は,カ ラーヒストグラムや輪郭の類似性など,比較的単純な指標が用いられることが多く,実環境での照明変動や複雑 背景に対する精度や頑健さなどの点で,必ずしも十分なものではなかった.これに対して本論文では,Haar-like 特徴を用いた AdaBoost 学習によるカスケード型識別器を仮説の評価に応用することで,頑健かつ高精度に人物 頭部を追跡する手法を提案する.更に,人物頭部の各方向に対応した識別器を複数準備し,パーティクルフィル(More)
For tracking a varying number of people using distributed cameras, it is important to deal with appearance/disappearance of people due to various causes including occlusions. We propose a method for tracking people with automatic initialization by integrating observations from distributed cameras and laser range scanners. In our method, the problem of(More)
In the present study, we describe a newly developed CNC lathe for realizing non-axisymmetric curved-surface-turning (NACS-turning). The newly developed lathe can be controlled along four axes (X1, X2, Z, and C) and has a three-linear-motor drive system for the X1, X2, and Z axes to control precise, high-speed motion. The tandem table unit, which moves along(More)
  • 1