摘要:
The present invention generally relates to a technology of detecting a stationary object in a static video, i.e., a video which is generated by a camera with a fixed view, such as a CCTV video. More specifically, the present invention relates to a technology of identifying a stationary object by use of inference information (Tensor, Activation Map) of Deep Learning object detector among objects which have been detected in a static video such as a CCTV video. According to the present invention, it is possible to lower the priority of stationary objects (e.g., parked vehicles) in processing a static video so that computing resource of the video analysis system can be effectively utilized in the process after object detection so as to enhance the performance of image analysis.
摘要:
The present invention generally relates to a technology of detecting a stationary object in a static video, i.e., a video which is generated by a camera with a fixed view, such as a CCTV video. More specifically, the present invention relates to a technology of identifying a stationary object by use of inference information (Tensor, Activation Map) of Deep Learning object detector among objects which have been detected in a static video such as a CCTV video. According to the present invention, it is possible to lower the priority of stationary objects (e.g., parked vehicles) in processing a static video so that computing resource of the video analysis system can be effectively utilized in the process after object detection so as to enhance the performance of image analysis.
摘要:
There is provided a polyamic acid resin composition, a method for preparing the same and a polyimide metal clad laminate using the same, in which the polyamic acid resin composition includes an epoxy compound represented by Chemical Formula 1 is defined in specification.
摘要:
A compression encoding method which, when performing data conversion using the fact that when data is converted into frequency components, the human eye is sensitive to components close to low frequency components and insensitive to components close to high frequency components, configures a filter in which importance is placed on DC data and its adjacent filter coefficients, calculates the cost for a conversion block which has undergone conversion and quantization, which is required for the data removal of compression encoding using the filter, determines data targets to be removed using the calculated cost and set reference values, and removes the corresponding data, thereby reducing the amount of data to be compressed and therefore increasing compression efficiency without influencing image quality.
摘要:
A circuit board includes; a reference common voltage generating circuit receiving an input voltage from an input node and outputting a reference common voltage to an output node, wherein the reference common voltage generating circuit includes a voltage distribution unit including a first resistor connected between the input node and the output node and a second resistor connected between the output node and a ground, and a sink current generating unit outputting a sink current to the output node, wherein the sink current is adjusted according to a resistance of a setting resistor, and a coupling line coupling the setting resistor and the reference common voltage generating circuit.