Information processing method, program, and information processing apparatus
The present technology relates to an information processing method, a program, and an information processing apparatus capable of analyzing a learning situation of a model using a neural network. In step S 1 , feature data that numerically represents a feature of a feature map generated from input data in a model using a neural network is generated. In step S 2 , analysis data based on the feature data of a plurality of the feature maps is generated. The present technology can be applied to, for example, a system that recognizes a vehicle in front of a vehicle.
1. An information processing method executed by processing circuitry, the method comprising:
generating feature data that numerically represents a feature of a feature map generated from input data in a model using a neural network;
generating analysis data based on the feature data of a plurality of feature maps; and
displaying the generated analysis data as feature data, based on feature maps generated from a same input data, arranged in numerical order of the feature maps in a first axis direction and feature data, based on feature maps corresponding to different input data, arranged in numerical order of the different input data in a second axis direction orthogonal to the first axis direction, wherein the generated analysis data is displayed as an image in which pixels represent values of the feature data of a given input data and a given feature map, and the image is indicative of a state of learning of the model using the neural network.
2. The information processing method according to claim 1 ,
wherein in the analysis data, the feature data of a plurality of the feature maps generated from a plurality of the input data is arranged in a predetermined hierarchy of the model that generates a plurality of the feature maps from one of the input data.
3. The information processing method according to claim 1 ,
wherein the feature data indicates a degree of dispersion of pixel values of the feature map.
4. The information processing method according to claim 3 , further comprising:
analyzing a learning situation of the model on a basis of a line in the first direction and a line in the second direction of the analysis data.
5. The information processing method according to claim 4 , further comprising:
setting a parameter for learning of the model on a basis of an analysis result of the learning situation of the model.
6. The information processing method according to claim 5 ,
wherein in the parameter setting, a regularization parameter for learning of the model is set on a basis of the number of lines in the second direction of the analysis data.
7. The information processing method according to claim 2 ,
wherein the feature data indicates a frequency distribution of pixel values of the feature map, and
in the analysis data, the feature data of a plurality of the feature maps generated from a plurality of the input data is three-dimensionally arranged in the hierarchy of the model.
8. The information processing method according to claim 1 ,
wherein the input data is image data, and
the model performs recognition processing of an object.
9. The information processing method according to claim 8 ,
wherein the model performs recognition processing of a vehicle.
10. The information processing method according to claim 9 ,
wherein the input data is image data representing a distribution of intensity of a received signal of a millimeter wave radar in a bird's-eye view.
11. The information processing method according to claim 10 ,
wherein the model converts the image data into an image in a camera coordinate system.
12. The information processing method according to claim 1 ,
wherein the analysis data includes the feature data satisfying a predetermined condition among the feature data of a plurality of the feature maps.
13. A non-transitory computer readable medium storing instructions that, when executed by processing circuitry, perform an information processing method comprising:
generating feature data that numerically represents a feature of a feature map generated from input data in a model using a neural network;
generating analysis data based on the feature data of a plurality of feature maps; and
displaying the generated analysis data as feature data, based on feature maps generated from a same input data, arranged in numerical order of the feature maps in a first axis direction and feature data, based on feature maps corresponding to different input data, arranged in numerical order of the different input data in a second axis direction orthogonal to the first axis direction, wherein the generated analysis data is displayed as an image in which pixels represent values of the feature data of a given input data and a given feature map, and the image is indicative of a state of learning of the model using the neural network.
14. An information processing apparatus comprising:
processing circuitry configured to:
generate feature data that numerically represents a feature of a feature map generated from input data in a model using a neural network;
generate analysis data based on the feature data of a plurality of feature maps; and
display the generated analysis data as feature data, based on feature maps generated from a same input data, arranged in numerical order of the feature maps in a first axis direction and feature data, based on feature maps corresponding to different input data, arranged in numerical order of the different input data in a second axis direction orthogonal to the first axis direction, wherein the generated analysis data is displayed as an image in which pixels represent values of the feature data of a given input data and a given feature map, and the image is indicative of a state of learning of the model using the neural network.