A Real-Time Human Detection System for Video
スポンサーリンク
概要
- 論文の詳細を見る
This work presents a real-time human detection system for VGA (Video Graphics Array, 640×480) video, which well suits visual surveillance applications. To achieve high running speed and accuracy, firstly we design multiple fast scalar feature types on the gradient channels, and experimentally identify that NOGCF (Normalized Oriented Gradient Channel Feature) has better performance with Gentle AdaBoost in cascaded classifiers. A confidence measure for cascaded classifiers is developed and utilized in the subsequent tracking stage. Secondly, we propose to use speedup techniques including a detector pyramid for multi-scale detection and channel compression for integral channel calculation respectively. Thirdly, by integrating the detector's discrete detected humans and continuous detection confidence map, we employ a two-layer tracking by detection algorithm for further speedup and accuracy improvement. Compared with other methods, experiments show the system is significantly faster with 20fps running speed in VGA video and has better accuracy as well.
- 2012-07-01
著者
-
LIN Xinggang
Department of Electronic Engineering, Tsinghua University
-
Lin Xinggang
Department Of Electronic Engineering Tsinghua University
-
Wang Guijin
Department Of Electronic Engineering Tsinghua University
-
ZENG Bobo
Department of Electronic Engineering, Tsinghua University
-
LIU Chunxiao
Department of Electronic Engineering, Tsinghua University
関連論文
- Online HOG Method in Pedestrian Tracking
- Robust Object Tracking via Combining Observation Models
- A Framework of Real Time Hand Gesture Vision Based Human-Computer Interaction
- Measuring Particles in Joint Feature-Spatial Space
- Kernel Based Image Registration Incorporating with Both Feature and Intensity Matching
- Real-Time Human Detection Using Hierarchical HOG Matrices
- DSP-Based Parallel Implementation of Speeded-Up Robust Features
- Non-rigid Object Tracking as Salient Region Segmentation and Association
- Partial Derivative Guidance for Weak Classifier Mining in Pedestrian Detection
- Drastic Anomaly Detection in Video Using Motion Direction Statistics
- High-Accuracy Sub-Pixel Registration for Noisy Images Based on Phase Correlation
- Stereo Matching Using Local Plane Fitting in Confidence-Based Support Window
- Implementation of Scale and Rotation Invariant On-Line Object Tracking Based on CUDA
- A Real-Time Human Detection System for Video
- An Interleaving Updating Framework of Disparity and Confidence Map for Stereo Matching
- Kernel-Based On-Line Object Tracking Combining both Local Description and Global Representation