Video conference appearance validation and remediation
An approach for preventing anomalous appearance characteristics of a user during a video conference. The approach receives a user's video feed from a video conference comprising the user's video feed. The approach analyzes the user's video feed based on a comparison with a reference appearance model associated with the user. The approach detects, based on the analysis, anomalous appearance characteristics associated with the user. The approach notifies the user of the anomalous appearance.
1 . A computer-implemented method for preventing anomalous appearance characteristics of a user during a video conference, the computer-implemented method comprising:
receiving, by one or more processors, a user's video feed from a video conference comprising the user's video feed and meeting data, wherein the meeting data comprises of emails, text messages and verbal communications;
creating, by a recurrent neural network (RNN), a reference appearance model of the user based on historical video appearances and user defined preferences;
training the reference appearance model, by the recurrent neural network (RNN), based on the user's video feed;
analyzing, by the recurrent neural network (RNN), the user's video feed based on a comparison with the reference appearance model associated with the user;
determining whether the video conference is a formal meeting or an informal meeting, wherein the formal meeting is a meeting that is associated with presenting a professional appearance for the user and is determined by executing a pre-meeting evaluation function;
in response to having determined that the video conference is the formal meeting, selecting a pre-meeting evaluation function to be performed on the user's video feed to validate the professional appearance of the user, where in the pre-meeting evaluation function further comprising:
comparing the user's video feed against the reference appearance model based on a predetermined threshold to determine a standard deviation of formality;
detecting, anomalous appearance characteristics associated with the user, wherein the anomalous appearance characteristics is characteristics that is outside the predetermined threshold of a professional image model;
notifying, by the one or more processors, the user of an anomalous appearance, wherein the anomalous appearance is an appearance that is outside a predetermined threshold of the professional image model; and
in response to having determined that the video conference is the informal meeting, detecting, by using a natural language processing (NLP) technique, informal data from the meeting data and removing of formal data associated with the reference appearance model.
2 . The computer-implemented method of claim 1 , wherein the reference appearance model is trained with a recurrent neural network.
3 . The computer-implemented method of claim 1 , further comprising:
executing, by the one or more processors, remedial actions comprising at least one of obfuscation of the anomalous appearance characteristics or feed manipulation of the user's feed to eliminate the anomalous appearance characteristics.
4 . The computer-implemented method of claim 1 , wherein the reference appearance model is trained based on at least one of verbal communication or messaging associated with the video conference.
5 . The computer-implemented method of claim 1 , wherein the reference appearance model is trained based on user reaction to notification of the anomalous appearance.
6 . The computer-implemented method of claim 1 , wherein the reference appearance model is selected based on attendees of a video conference.
7 . A computer program product for preventing anomalous appearance characteristics of a user during a video conference, the computer program product comprising:
one or more non-transitory computer readable storage media and program instructions stored on the one or more non-transitory computer readable storage media, the program instructions comprising:
program instructions to receive a user's video feed from a video conference comprising the user's video feed and meeting data, wherein the meeting data comprises of emails, text messages and verbal communications;
program instructions to create, by a recurrent neural network (RNN), a reference appearance model of the user based on historical video appearances and user defined preferences;
program instructions to train the reference appearance model, by the recurrent neural network (RNN), based on the user's video feed;
program instructions to analyze, by the recurrent neural network (RNN), the user's video feed based on a comparison with the reference appearance model associated with the user;
program instructions to determine whether the video conference is a formal meeting or an informal meeting, wherein the formal meeting is a meeting that is associated with presenting a professional appearance for the user and is determined by executing a pre-meeting evaluation function;
in response to having determined that the video conference is the formal meeting, program instructions to select a pre-meeting evaluation function to be performed on the user's video feed to validate the professional appearance of the user, wherein the pre-meeting evaluation function further comprising:
program instructions to compare the user's video feed against the reference appearance model based on a predetermined threshold to determine a standard deviation of formality;
program instructions to detect, anomalous appearance characteristics associated with the user, wherein the anomalous appearance characteristics is characteristics that is outside the predetermined threshold of a professional image model;
program instructions to notify the user of the anomalous appearance, wherein the anomalous appearance is an appearance that is outside a predetermined threshold of the professional image model; and
in response to having determined that the video conference is the informal meeting, program instructions to detect, by using a natural language processing (NLP) technique, informal data from the meeting data and removing of formal data associated with the reference appearance model.
8 . The computer program product of claim 7 , wherein the reference appearance model is trained with a recurrent neural network.
9 . The computer program product of claim 7 , further comprising:
program instructions to execute remedial actions comprising at least one of obfuscation of the anomalous appearance characteristics or feed manipulation of the user's feed to eliminate the anomalous appearance characteristics.
10 . The computer program product of claim 7 , wherein the reference appearance model is trained based on at least one of verbal communication or messaging associated with the video conference.
11 . The computer program product of claim 7 , wherein the reference appearance model is trained based on user reaction to notification of the anomalous appearance.
12 . The computer program product of claim 7 , wherein the reference appearance model is selected based on attendees of a video conference.
13 . A computer system for preventing anomalous appearance characteristics of a user during a video conference, the computer system comprising:
one or more computer processors;
one or more non-transitory computer readable storage media; and
program instructions stored on the one or more non-transitory computer readable storage media, the program instructions comprising:
program instructions to receive a user's video feed from a video conference comprising the user's video feed and meeting data, wherein the meeting data comprises of emails, text messages and verbal communications;
program instructions to create, by a recurrent neural network (RNN), a reference appearance model of the user based on historical video appearances and user defined preferences;
program instructions to train the reference appearance model, by the recurrent neural network (RNN), based on the user's video feed;
program instructions to analyze, by the recurrent neural network (RNN), the user's video feed based on a comparison with the reference appearance model associated with the user;
program instructions to determine whether the video conference is a formal meeting or an informal meeting, wherein the formal meeting is a meeting that is associated with presenting a professional appearance for the user and is determined by executing a pre-meeting evaluation function;
in response to having determined that the video conference is the formal meeting, program instructions to select a pre-meeting evaluation function to be performed on the user's video feed to validate the professional appearance of the user, wherein the pre-meeting evaluation function further comprising:
program instructions to compare the user's video feed against the reference appearance model based on a predetermined threshold to determine a standard deviation of formality;
program instructions to detect, anomalous appearance characteristics associated with the user, wherein the anomalous appearance characteristics is characteristics that is outside the predetermined threshold of a professional image model;
program instructions to notify the user of the anomalous appearance, wherein the anomalous appearance is an appearance that is outside a predetermined threshold of the professional image model; and
in response to having determined that the video conference is the informal meeting, program instructions to detect, by using a natural language processing (NLP) technique, informal data from the meeting data and removing of formal data associated with the reference appearance model.
14 . The computer system of claim 13 , wherein the reference appearance model is trained with a recurrent neural network.
15 . The computer system of claim 13 , further comprising:
program instructions to execute remedial actions comprising at least one of obfuscation of the anomalous appearance characteristics or feed manipulation of the user's feed to eliminate the anomalous appearance characteristics.
16 . The computer system of claim 13 , wherein the reference appearance model is trained based on user reaction to notification of the anomalous appearance and based on at least one of verbal communication or messaging associated with the video conference.
17 . The computer system of claim 13 , wherein the reference appearance model is selected based on attendees of a video conference.