Information processing apparatus, information processing method, and storage medium

1. An information processing apparatus comprising:

one or more memories storing instructions; and

one or more processors that, upon execution of the stored instructions, are configured to:

acquire, from each of a plurality of partial regions obtained by dividing an input image, likelihood information indicating a likelihood indicating certainty of presence of a particular object;

determine, based on the likelihood information, a region where the likelihood is greater than or equal to a first predetermined value among the plurality of partial regions, as a region where a threshold is to be adjusted to be lower; and

estimate a number of particular objects by counting the likelihood with respect to each of the partial regions by excluding a likelihood less than the threshold among the likelihoods included in the partial regions from counting targets.

2. The information processing apparatus according to claim 1 , wherein determines a region where there are a certain number or more of likelihoods greater than a second predetermined value among the likelihoods included in the partial regions is determined, as a region where the threshold in the partial regions is to be adjusted to be lower.

3. The information processing apparatus according to claim 1 , wherein the one or more processors are further configured to generate a plurality of partial regions from the input image.

4. The information processing apparatus according to claim 3 ,

wherein based on the likelihood information, the one or more processors generates a partial region of interest from the input image, and

wherein based on the likelihood information included in the partial region of interest, the one or more processors determines a region where the threshold is to be adjusted.

5. The information processing apparatus according to claim 4 ,

wherein based on the likelihood information, the one or more processors generates any of the partial regions where the number of the likelihoods is greater than a predetermined number, as the partial region of interest, and

wherein the one or more processors determine the partial region of interest as a region where the threshold is to be adjusted to be lower.

6. The information processing apparatus according to claim 4 ,

wherein based on the likelihood information, the one or more processors generates a region included in a predetermined range about a likelihood greater than or equal to a predetermined value included in the partial regions, as the partial region of interest, and

wherein the one or more processors determine the partial region of interest as a region where the threshold is to be adjusted to be lower.

7. The information processing apparatus according to claim 3 , wherein in a case where the likelihood information changes between frames of a moving image, the one or more processors update the partial regions.

8. The information processing apparatus according to claim 3 , wherein the partial regions are generated according to an instruction from a user.

9. The information processing apparatus according to claim 1 ,

wherein the likelihood information is acquired as a result of inputting the input image to a trained model that estimates positions of the particular objects, and

wherein the trained model outputs the likelihood information regarding each of the plurality of partial regions.

10. The information processing apparatus according to claim 1 , wherein the one or more processors are further configured to set a predetermined threshold with respect to each of the plurality of partial regions,

wherein in a case where there are a certain number or more of likelihoods greater than or equal to a predetermined value in the partial region, the one or more processors determines the partial region as a region where the threshold is to be adjusted to be lower than the predetermined threshold.

11. The information processing apparatus according to claim 1 , wherein according to the number of likelihoods greater than or equal to a predetermined value that are included in each of the partial regions, the one or more processors determine the threshold.

12. The information processing apparatus according to claim 1 , wherein the one or more processors adjust the threshold in any of the partial regions where there is not a likelihood greater than or equal to a predetermined value, to be higher than the threshold in any of the partial regions where there is a likelihood greater than or equal to the predetermined value.

13. The information processing apparatus according to claim 12 , wherein according to an instruction to make an automatic determination from a user or an indication of a threshold from the user, the one or more processors set the threshold with respect to each of the partial regions.

14. The information processing apparatus according to claim 1 ,

wherein the particular objects are people, and

wherein the one or more processors estimate the number of people included in the input image.

15. The information processing apparatus according to claim 1 , wherein the one or more processors estimate a sum total value of likelihoods included in the likelihood information, each of which is greater than the threshold, as the number of the particular objects.

16. An information processing method comprising:

acquiring, from each of a plurality of partial regions obtained by dividing an input image, likelihood information indicating a likelihood indicating certainty of presence of a particular object;

based on the likelihood information, determining a region where the likelihood is greater than or equal to a first predetermined value among the plurality of partial regions, as a region where a threshold is to be adjusted to be lower; and

estimating a number of particular objects by counting the likelihood with respect to each of the partial regions by excluding a likelihood less than the threshold among the likelihoods included in the partial regions from counting targets.

17. A non-transitory computer-readable storage medium storing a program for causing a computer to execute an information processing method comprising:

acquiring, from each of a plurality of partial regions obtained by dividing an input image, likelihood information indicating a likelihood indicating certainty of presence of a particular object;

based on the likelihood information, determining a region where the likelihood is greater than or equal to a first predetermined value among the plurality of partial regions, as a region where a threshold is to be adjusted to be lower; and

estimating a number of particular objects by counting the likelihood with respect to each of the partial regions by excluding a likelihood less than the threshold among the likelihoods included in the partial regions from counting targets.