[PDF][PDF] A Preliminary Study on Human Attribute Classification in Thermal Image
2015 年電子情報通信学会総合大会講演論文集, 2015•vislab.is.i.nagoya-u.ac.jp
Nowadays, security surveillance systems are implemented in various location. A basic
surveillance system usually employs camera that works in visible spectrum, where human
can see. During nighttime or in poorly lit areas, thermal camera can be a better option, as it
can capture images without relying on illumination. Figure 1 shows the examples of images
taken in different spectra. Regardless of the camera choice, a person captured by the
surveillance camera is not usually easy to be identified. It is also quite common to see …
surveillance system usually employs camera that works in visible spectrum, where human
can see. During nighttime or in poorly lit areas, thermal camera can be a better option, as it
can capture images without relying on illumination. Figure 1 shows the examples of images
taken in different spectra. Regardless of the camera choice, a person captured by the
surveillance camera is not usually easy to be identified. It is also quite common to see …
Nowadays, security surveillance systems are implemented in various location. A basic surveillance system usually employs camera that works in visible spectrum, where human can see. During nighttime or in poorly lit areas, thermal camera can be a better option, as it can capture images without relying on illumination. Figure 1 shows the examples of images taken in different spectra.
Regardless of the camera choice, a person captured by the surveillance camera is not usually easy to be identified. It is also quite common to see people wearing accessories such as masks, glasses and hats, making the identification process harder. Therefore, it is useful to classify these visual attributes to find a certain person in surveillance footage.
vislab.is.i.nagoya-u.ac.jp
以上显示的是最相近的搜索结果。 查看全部搜索结果