Dynamic adaptation of feature identification and annotation

A system, comprising:

• a detector, associated with labeled data, comprising a non-transitory computer readable medium storing first instructions that, when executed, cause the system to perform one or more operations comprising:
• accessing input data including an image;
• and
• identifying a feature in the image and an associated candidate tag based on items of interest, and determine an associated accuracy metric based on the feature and the candidate tag, wherein the items of interest include one or more features and one or more associated tags;
• a scheduler comprising a non-transitory computer readable medium storing second instructions that, when executed, cause the system to selectively obtain feedback, having a feedback accuracy, about the candidate tag from a set of sources based on the accuracy metric and a feedback threshold;
• a revision module comprising a non-transitory computer readable medium storing third instructions that, when executed, cause the system to generate a revised tag based on the feedback when the feedback indicates the candidate tag is incorrect;
• a presentation module comprising a non-transitory computer readable medium storing fourth instructions that, when executed, cause the system to present one of:
• a first result based on the accuracy metric and a detection threshold;
• and a second result based on the feedback accuracy and the detection threshold;
• and
• a training module comprising a non-transitory computer readable medium storing fifth instructions that, when executed, cause the system to update, based on a quality metric, the labeled data to include the input data, the feature and one of the candidate tag and the revised tag, wherein the updated labeled data are to be used to retrain the detector.

A non-transitory computer-readable storage medium for use in conjunction with a system, the computer-readable storage medium storing a program module, wherein, when executed by the system, the program module causes the system to selectively provide a first result or a second result by performing one or more operations comprising:

• accessing input data including an image;
• identifying, using a detector associated with labeled data, a feature in the image and an associated candidate tag based on items of interest, and determining an associated accuracy metric based on the feature and the candidate tag, wherein the items of interest include one or more features and one or more associated tags;
• selectively obtaining, using a scheduler, feedback, having a feedback accuracy, about the candidate tag from a set of sources based on the accuracy metric and a feedback threshold;
• generating a revised tag based on the feedback when the feedback indicates the candidate tag is incorrect;
• presenting one of:
• the first result based on the accuracy metric and a detection threshold;
• and
• the second result based on the feedback accuracy and the detection threshold;
• and
• updating, based on a quality metric, the labeled data to include the input data, the feature and one of the candidate tag and the revised tag, wherein the updated labeled data are to be used to retrain the detector.

A method for selectively presenting a first result or a second result, the method comprising:

• by a system:
• accessing input data including an image;
• identifying, using a detector associated with labeled data, a feature in the image and an associated candidate tag based on items of interest, and determining an associated accuracy metric based on the feature and the candidate tag, wherein the items of interest include one or more features and one or more associated tags;
• selectively obtaining, using a scheduler, feedback, having a feedback accuracy, about the candidate tag from a set of sources based on the accuracy metric and a feedback threshold;
• generating a revised tag based on the feedback when the feedback indicates the candidate tag is incorrect;
• presenting one of:
• a first result based on the accuracy metric and a detection threshold;
• and a second result based on the feedback accuracy and the detection threshold;
• and
• updating, based on a quality metric, the labeled data to include the input data, the feature and one of the candidate tag and the revised tag, wherein the updated labeled data are to be used to retrain the detector.