Generating concept images of human poses using machine learning models

1. A computer-implemented method comprising:

identifying one or more events from input data by applying a machine learning recognition model to the input data, wherein said identifying comprises (i) detecting multiple entities from the input data, (ii) determining one or more behavioral relationships among the multiple entities in the input data, and (iii) generating, via the machine learning recognition model, a vector representation of one or more human poses related to the identified events;

generating, using a machine learning interpretability model and the identified events, one or more images illustrating one or more human poses related to the identified events, wherein said generating the one or more images comprises converting the vector representation to one or more human pose representations over one or more stick figure images;

outputting the one or more generated images to at least one user; and

updating the machine learning recognition model based at least in part on (i) the one or more generated images and (ii) input from the at least one user;

wherein the method is carried out by at least one computing device.

2. The computer-implemented method of claim 1 , comprising:

training the machine learning recognition model using data pertaining to human pose structures and image texture information.

3. The computer-implemented method of claim 1 , wherein the input data comprise image data.

4. The computer-implemented method of claim 1 , wherein the input data comprise text data.

5. The computer-implemented method of claim 1 , wherein the input data comprise multimodal data.

6. The computer-implemented method of claim 1 , wherein the one or more images comprise one or more stick figure images.

7. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computing device to cause the computing device to:

identify one or more events from input data by applying a machine learning recognition model to the input data, wherein said identifying comprises (i) detecting multiple entities from the input data, (ii) determining one or more behavioral relationships among the multiple entities in the input data, and (iii) generating, via the machine learning recognition model, a vector representation of one or more human poses related to the identified events;

generate, using a machine learning interpretability model and the identified events, one or more images illustrating one or more human poses related to the identified events, wherein said generating the one or more images comprises converting the vector representation to one or more human pose representations over one or more stick figure images;

output the one or more generated images to at least one user; and

update the machine learning recognition model based at least in part on (i) the one or more generated images and (ii) input from the at least one user.

8. The computer program product of claim 7 , wherein the input data comprise one or more of image data, text data, and multimodal data.

9. A system comprising:

a memory; and

at least one processor operably coupled to the memory and configured for:

identifying one or more events from input data by applying a machine learning recognition model to the input data, wherein said identifying comprises (i) detecting multiple entities from the input data, (ii) determining one or more behavioral relationships among the multiple entities in the input data, and (iii) generating, via the machine learning recognition model, a vector representation of one or more human poses related to the identified events;

generating, using a machine learning interpretability model and the identified events, one or more images illustrating one or more human poses related to the identified events, wherein said generating the one or more images comprises converting the vector representation to one or more human pose representations over one or more stick figure images;

outputting the one or more generated images to at least one user; and

updating the machine learning recognition model based at least in part on (i) the one or more generated images and (ii) input from the at least one user.

10. A computer-implemented method comprising:

training a machine learning recognition model using data pertaining to human pose structures and image texture information;

identifying one or more events from input data by applying the machine learning recognition model to the input data, wherein said identifying comprises (i) detecting multiple entities from the input data, (ii) determining one or more behavioral relationships among the multiple entities in the input data, and (iii) generating, via the machine learning recognition model, a vector representation of one or more human poses related to the identified events;

generating, using a machine learning interpretability model and the identified events, one or more images illustrating one or more human poses related to the identified events, wherein said generating the one or more images comprises converting the vector representation to one or more human pose representations over one or more stick figure images;

outputting, via a dialog system, the one or more generated images to at least one human user;

automatically producing at least one machine-originated image representative of the one or more generated images based at least in part on feedback from the at least one human user;

updating the machine learning recognition model based at least in part on the at least one machine-originated image;

wherein the method is carried out by at least one computing device.

11. The computer-implemented method of claim 10 , wherein the input data comprise image data.

12. The computer-implemented method of claim 10 , wherein the input data comprise text data.

13. The computer-implemented method of claim 10 , wherein the input data comprise multimodal data.

14. The computer-implemented method of claim 10 , wherein the one or more images comprise one or more stick figure images.