Information processing apparatus, information processing method, and program

Abstract

To provide an information processing apparatus, an information processing method, and a program allowing a user to efficiently check the validity of a result of learning of an image classification AI and take measures to improve the accuracy of the image classification AI.SOLUTION: A CPU 201 of an information processing apparatus 101 infers a prediction label from an input image by using an image classification AI and calculates a heat map, and displays a prediction basis image in which the input image, prediction label, and heat map are superimposed on the input image and a correct answer label as one screen in a comparable manner on a result analysis screen of a display 212. Subsequently, when a user depresses at least one of a learning button 405, a processing button 406, and a deletion button 407 on the result analysis screen, and then depresses an export button 408, the CPU 201 updates learning data by using the input image so as to perform re-learning of the image classification AI.SELECTED DRAWING: Figure 4

Claims

1. a display control means for controlling to display an input image, a predicted label which is a result of inference by a trained model for the input image, a grounds image in which a feature region which is the grounds for the inference by the trained model is superimposed on the input image, and a correct answer label which indicates the correct answer of the inference for the input image in a corresponding manner; A receiving means for receiving an instruction to update learning data for re-learning the trained model using the input image; an update means for updating the learning data by using the input image when the instruction is accepted by the acceptance means; An information processing device comprising:

2. 2 . The information processing apparatus according to claim 1 , wherein the accepting means includes first accepting means for accepting an instruction to use the input image as the learning data after processing the input image.

3. The information processing device according to claim 2, characterized in that, when an instruction is received from the first receiving means after the input image is acquired as the learning data, the updating means replaces the input image in the learning data with the processed input image, thereby updating the learning data.

4. The information processing device according to claim 2, characterized in that, when an instruction is received from the first receiving means after the input image is acquired as evaluation data for evaluating the trained model, the updating means moves the processed input image from the evaluation data to the training data, thereby updating the training data and the evaluation data.

5. 5 . The information processing apparatus according to claim 1 , wherein the accepting means includes a second accepting means for accepting an instruction not to use the input image as the learning data.

6. The information processing device according to claim 5, characterized in that, when the update means receives an instruction from the second reception means after acquiring the input image as the learning data, it deletes the input image from the learning data and updates the learning data.

7. The information processing device according to claim 5, characterized in that, when the update means receives an instruction from the second reception means after acquiring the input image as evaluation data for evaluating the trained model, the update means deletes the input image from the evaluation data and updates the evaluation data.

8. The reception means includes a third reception means for receiving, when the input image is acquired as evaluation data for evaluating the trained model, an instruction to use the input image as the training data; The information processing device according to any one of claims 1 to 7, characterized in that, when an instruction is received from the third receiving means, the update means moves the input image from the evaluation data to the learning data, thereby updating the learning data and the evaluation data.

9. A display control process in which a display control means of an information processing device controls to display an input image, a predicted label based on a trained model for the input image, a prediction basis image in which a feature region that is the basis for the prediction by the trained model is superimposed on the input image, and a correct answer label indicating a correct answer for the input image in association with each other; A receiving step in which a receiving means of the information processing device receives an instruction to update learning data for re-learning the trained model using the input image; an updating step of updating the learning data by using the input image when the updating means of the information processing device accepts the instruction by the accepting means of the information processing device; An information processing method comprising:

10. A computer-executable program that causes a computer to function as each of the means of the information processing apparatus according to any one of claims 1 to 8.

Images