Thermo-key: human region segmentation from video

@article{Yasuda2004ThermokeyHR,
  title={Thermo-key: human region segmentation from video},
  author={Kazutaka Yasuda and Takeshi Naemura and Hiroshi Harashima},
  journal={IEEE Computer Graphics and Applications},
  year={2004},
  volume={24},
  pages={26-30}
}
We focus on the segmentation of human regions, an area that has had a significant amount of research. [] Key Method A color and thermal combination camera measures the temperature distribution of its field of view. This system segments the human region from the video sequence - captured by the color camera in real time - with high robustness against lighting and background conditions.

Figures and Tables from this paper

Live Video Segmentation in Dynamic Backgrounds Using Thermal Vision
TLDR
A new technique for live video segmentation of human regions from dynamic backgrounds with the employing of the thermal cue which proves to be very effective when fused with color and a new speed-up GraphCut algorithm by combining with the Bayesian estimation.
Thermal Infrared and Visible Spectrum Fusion for Multi-modal Video Analysis
TLDR
The challenge of multimodal fusion is to obtain optimal benefits from the joint use of multiple modalities, namely thermal infrared and visible spectrum video in this project.
User region extraction from dynamic projected background for a virtual environment system
TLDR
A background subtraction method is proposed that segments dynamic projected background in a rear-projection-based virtual environment system and separate user region from dynamic projected Background by simple subtraction between camera images and the computed background.
VisMerge: Light Adaptive Vision Augmentation via Spectral and Temporal Fusion of Non-visible Light
TLDR
VisMerge is a framework that combines a thermal imaging head mounted display (HMD) and algorithms that temporally and spectrally merge video streams of different light bands into the same field of view, and shows that freedom of user movement is significantly increased with visualizations engaged.
Stripe Chromakey Using Active Contour Models
正会員 山 下 淳 y;yy , 縣 弘 樹 y , 正会員 金 子 透 y Abstract We propose a new region extraction and image composition method with chromakey using a two-tone striped background and active contour models. A problem
Superimposing Thermal-Infrared Data on 3D Structure Reconstructed by RGB Visual Odometry
TLDR
This paper proposes a method for estimating the scale of the point cloud corresponding to the ex-trinsic parameters between both cameras by matching depth images recovered from the RGB camera and the thermal-infrared camera based on mutual information.
Evaluation of Synchronization Accuracy between High Speed Cameras in Infrared and Visible Spectrums
TLDR
A new approach for evaluating synchronization accuracy between infrared (thermographic) camera and visible spectrum high speed camera is presented, based on angle differences in the specially designed rotating marker on corresponding frames.
Adaptive detection and tracking using multimodal information
TLDR
This thesis describes work on fusing data from multiple sources of information, and explores a new paradigm in dynamic parameter selection, by selecting thresholds for object detection to maximise agreement between pairs of sources.
Stripe Chromakey Using Active Contour Models
TLDR
A new region extraction and image composition method with chromakey using a two-tone striped background and active contour models to extract foreground regions whose colors are same with those of the background.
...
...

References

SHOWING 1-10 OF 17 REFERENCES
Thermo-Key [vision camera]
The Thermo-key real-time system segments human regions from video sequences using thermal information This approach is robust and easy to use in comparison with chroma-key systems. The movie contains
Background modeling for segmentation of video-rate stereo sequences
TLDR
A method of statistical background modeling for stereo sequences that improves the reliability and sensitivity of segmentation in the presence of object clutter is presented, and the dynamic version of the method, called gated background adaptation, can reliably learn background statistics in the absence of corrupting foreground motion.
Video background replacement without a blue screen
  • R. J. QianM. Sezan
  • Computer Science
    Proceedings 1999 International Conference on Image Processing (Cat. 99CH36348)
  • 1999
TLDR
An algorithm is presented for replacing video background without a physical blue screen by compiling a probability map which contains the likelihood for each pixel to be classified into the foreground or background.
Fast Lighting Independent Background Subtraction
TLDR
A simple method of fast background subtraction based upon disparity verification that is invariant to arbitrarily rapid run-time changes in illumination that is easily implemented in real-time on conventional hardware.
Background estimation and removal based on range and color
TLDR
A background estimation method based on a multidimensional (range and color) clustering at each image pixel is described and demonstrated and important implementation issues such as treatment of shadows and low confidence measurements are discussed in detail.
A Survey of Computer Vision-Based Human Motion Capture
A comprehensive survey of computer vision-based human motion capture literature from the past two decades is presented. The focus is on a general overview based on a taxonomy of system
Log-Tracker: an Attribute-Based Approach to Tracking Human Body Motion
TLDR
Experimental results from two gymnastics sequences indicate that the new Log-Tracker system is able to track the body parts even when they are occluded for a short period of time, and occlusion of body parts is termed a forking condition.
Axi-vision camera: a three-dimensional camera
TLDR
A novel 3D camera named Axi-vision camera was developed that can acquire both color and distance information of objects and its application to TV program production, such as replacing the image of an object at a particular distance by another, was demonstrated.
W 4 : Who ? When ? Where ? What ? A Real Time System for Detecting and Tracking
TLDR
The computational models employed by W 4 to detect and track people and their parts are described to allow it to determine types of interactions between people and objects, and to overcome the inevitable errors and ambiguities that arise in dynamic image analysis.
Tracking people with twists and exponential maps
  • C. BreglerJitendra Malik
  • Computer Science
    Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231)
  • 1998
TLDR
This paper demonstrates a new visual motion estimation technique that is able to recover high degree-of-freedom articulated human body configurations in complex video sequences, and is the first computer vision based system able to process such challenging footage and recover complex motions with such high accuracy.
...
...