Processing method of learning face recognition by artificial intelligence module

1. A processing method of face recognition, comprising steps of:

extracting embedding feature information from a face image;

outputting a recognition result of a face recognition according to the embedding feature information, wherein the recognition result comprises a recognized name and embedding feature distance information;

determining whether the recognized name is in a list or not;

if the recognized name is in the list, performing a removal checking step for determining whether to remove the recognition result based on the embedding feature distance information;

if determining that the recognition result is not to be removed, displaying the recognized name;

if determining that the recognition result is to be removed, displaying a negative prompt; and

dynamically and instantly providing a feedback and updating a recognition method for the face recognition.

2. The processing method of claim 1 , wherein the removal checking step comprises:

determining whether the recognition result is ambiguous or not;

if determining that the recognition result is ambiguous, providing a plurality of candidate names for a user to select a correct name therefrom; and

if the correct name is different from the recognized name, updating a face information database according to the correct name and the embedding feature distance information.

3. The processing method of claim 2 , wherein the embedding feature distance information comprises an embedding feature distance, and the step of determining whether the recognition result is ambiguous or not comprises:

if the embedding feature distance is less than a low threshold value, determining that the recognition result is not to be removed;

if the embedding feature distance is greater than a high threshold value, determining that the recognition result is to be removed, wherein the high threshold value is greater than the low threshold value; and

if the embedding feature distance is between the low threshold value and the high threshold value, determining that the recognition result is ambiguous.

4. The processing method of claim 2 , wherein the step of providing the plurality of the candidate names comprises:

calculating all face embedding feature information stored in the face information database to obtain a plurality of face distance information;

sorting the face distance information;

selecting from the face distance information to obtain N sets of the face distance information with minimum distances as candidate distance information;

voting names corresponding to the candidate distance information; and

displaying a name voting result, wherein the name voting result comprises the candidate names.

5. The processing method of claim 2 , wherein the step of updating the face information database comprises:

removing a least approximate record with respect to the correct name from the face information database; and

adding a new record with respect to the correct name to the face information database according to the embedding feature distance information.

6. The processing method of claim 2 , wherein the removal checking step further comprising:

if the correct name is identical to the recognized name, not updating the face information database and displaying the recognized name.

7. The processing method of claim 1 , further comprising:

if determining that the recognized name is not in the list, performing an update checking step for adding a name information to the list.

8. The processing method of claim 1 , wherein a neural network of an AI module is used to extract the embedding feature information from the face image, and a classifier of the AI module is used to output the recognition result according to the embedding feature information.

9. A processing method of face recognition, comprising steps of:

extracting embedding feature information from a face image;

outputting a recognition result of a face recognition according to the embedding feature information, wherein the recognition result comprises a recognized name and embedding feature distance information;

determining whether the recognized name is in a list or not;

if determining that the recognized name is not in the list, performing an update checking step for adding a name information to the list;

if the recognized name is in the list, performing a removal checking step for determining whether to remove the recognition result based on the embedding feature distance information;

if determining that the recognition result is not to be removed, displaying the recognized name;

if determining that the recognition result is to be removed, displaying a negative prompt; and

dynamically and instantly providing a feedback and updating a recognition method for the face recognition;

wherein, the removal checking step comprises:

determining whether the recognition result is ambiguous or not;

if determining that the recognition result is ambiguous, providing a plurality of candidate names for a user to select a correct name therefrom;

if the correct name is identical to the recognized name, not updating a face information database and displaying the recognized name; and

if the correct name is different from the recognized name, updating the face information database according to the correct name and the embedding feature distance information;

wherein, the embedding feature distance information comprises an embedding feature distance, and the step of determining whether the recognition result is ambiguous or not comprises:

if the embedding feature distance is less than a low threshold value, determining that the recognition result is not to be removed;

if the embedding feature distance is greater than a high threshold value, determining that the recognition result is to be removed, wherein the high threshold value is greater than the low threshold value; and

if the embedding feature distance is between the low threshold value and the high threshold value, determining that the recognition result is ambiguous;

wherein, the step of providing the plurality of the candidate names comprises:

calculating all face embedding feature information stored in the face information database to obtain a plurality of face distance information;

sorting the face distance information;

selecting from the face distance information to obtain N sets of the face distance information with minimum distances as candidate distance information;

voting names corresponding to the candidate distance information; and

displaying a name voting result, wherein the name voting result comprises the candidate names; and

wherein, the step of updating the face information database comprises:

removing a least approximate record with respect to the correct name from the face information database; and

adding a new record with respect to the correct name to the face information database according to the embedding feature distance information.

10. The processing method of claim 9 , wherein a neural network of an AI module is used to extract the embedding feature information from the face image, and a classifier of the AI module is used to output the recognition result according to the embedding feature information.

11. An electronic device for performing a processing method of face recognition,

wherein the electronic device comprises steps of:

extracting embedding feature information from a face image;

outputting a recognition result of a face recognition according to the embedding feature information, wherein the recognition result comprises a recognized name and embedding feature distance information;

determining whether the recognized name is in a list or not;

if the recognized name is in the list, performing a removal checking step for determining whether to remove the recognition result based on the embedding feature distance information;

if determining that the recognition result is not to be removed, displaying the recognized name;

if determining that the recognition result is to be removed, displaying a negative prompt; and

dynamically and instantly providing a feedback and updating a recognition method for the face recognition.

12. The electronic device of claim 11 , wherein the removal checking step comprises:

determining whether the recognition result is ambiguous or not;

if determining that the recognition result is ambiguous, providing a plurality of candidate names for a user to select a correct name therefrom; and

if the correct name is different from the recognized name, updating a face information database according to the correct name and the embedding feature distance information.

13. The electronic device of claim 12 , wherein the embedding feature distance information comprises an embedding feature distance, and the step of determining whether the recognition result is ambiguous or not comprises:

if the embedding feature distance is less than a low threshold value, determining that the recognition result is not to be removed;

if the embedding feature distance is greater than a high threshold value, determining that the recognition result is to be removed, wherein the high threshold value is greater than the low threshold value; and

if the embedding feature distance is between the low threshold value and the high threshold value, determining that the recognition result is ambiguous.

14. The electronic device of claim 12 , wherein the step of providing the plurality of the candidate names comprises:

calculating all face embedding feature information stored in the face information database to obtain a plurality of face distance information;

sorting the face distance information;

selecting from the face distance information to obtain N sets of the face distance information with minimum distances as candidate distance information;

voting names corresponding to the candidate distance information; and

displaying a name voting result, wherein the name voting result comprises the candidate names.

15. The electronic device of claim 12 , wherein the step of updating the face information database comprises:

removing a least approximate record with respect to the correct name from the face information database; and

adding a new record with respect to the correct name to the face information database according to the embedding feature distance information.

16. The electronic device of claim 12 , wherein the removal checking step further comprising:

if the correct name is identical to the recognized name, not updating the face information database and displaying the recognized name.

17. The electronic device of claim 11 , wherein the processing method further comprises:

if determining that the recognized name is not in the list, performing an update checking step for adding a name information to the list.

18. The electronic device of claim 11 , further comprising:

an AI module comprising a neural network for extracting the embedding feature information from the face image, and a classifier for outputting the recognition result according to the embedding feature information.