Methods and systems for determining accuracy of sport-related information extracted from digital video frames
A computing system determines accuracy of sport-related information extracted from a time sequence of digital video frames that represent a sport event, the extracted sport-related information including an attribute that changes over the time sequence. The computing system (a) detects, based on the extracted sport-related information, a pattern of change of the attribute over the time sequence and (b) makes a determination of whether the detected pattern is an expected pattern of change associated with the sport event. If the determination is that the detected pattern is the expected pattern, then, responsive to making the determination, the computing system takes a first action that corresponds to the sport-related information being accurate. Whereas, if the determination is that the detected pattern is not the expected pattern, then, responsive to making the determination, the computing system takes a second action that corresponds to the sport-related information being inaccurate.
1. A method for determining accuracy of sport-related information extracted from a time sequence of digital video frames that represent a sport event, wherein the extracted sport-related information includes an attribute that changes over the time sequence, the method comprising:
detecting by a computing system, based on the extracted sport-related information, a pattern of change of the attribute over the time sequence, wherein the computing system comprises a processing unit and the computing system has access to mapping data that indicates, for each of a plurality of sport events, an expected pattern of change of the attribute;
identifying, by the computing system, the sport event represented by the digital video frames;
determining, by the computing system, the expected pattern of change of the attribute indicated by the mapping data for the identified sport event;
making a determination, by the computing system, of whether the detected pattern of change of the attribute over the time sequence is the expected pattern of change of the attribute indicated by the mapping data for the identified sport event;
if the determination is that the detected pattern of change of the attribute over the time sequence is the expected pattern of change of the attribute indicated by the mapping data of the identified sport event, then, responsive to making the determination, the computing system taking a first action that corresponds to the sport-related information being accurate; and
if the determination is that the detected pattern of change of the attribute over the time sequence is not the expected pattern of change of the attribute indicated by the mapping data for the identified sport event, then, responsive to making the determination, the computing system taking a second action that corresponds to the sport-related information being inaccurate.
2. The method of claim 1 , wherein the attribute is a score associated with the sport event.
3. The method of claim 1 , wherein the attribute is a time depicted by the digital video frames in accordance with the sport event.
4. The method of claim 1 ,
wherein the sport-related information includes (i) a first instance of the attribute as extracted from a first digital video frame of the digital video frames and (ii) a second instance of the attribute as extracted from a second digital video frame of the digital video frames, and
wherein detecting the pattern of change of the attribute comprises determining at least a difference between the first and second instances of the attribute.
5. The method of claim 4 , wherein the second digital video frame immediately follows the first digital video frame in the time sequence.
6. The method of claim 4 , further comprising:
determining, by the computing system, a time period between the first and second digital video frames in the time sequence,
wherein making the determination comprises making a determination of whether the determined difference corresponds to an expected difference between the first and second instances of the attribute over the determined time period.
7. The method of claim 1 ,
wherein detecting the pattern of change of the attribute comprises detecting a rate of increase of the attribute,
wherein making the determination comprises making a determination of whether the detected rate of increase of the attribute is (i) below a threshold rate of increase of the attribute or rather (ii) at or above the threshold rate of increase of the attribute,
wherein the computing system takes the first action if the determination is that the detected rate of increase of the attribute is below the threshold rate of increase of the attribute, and
wherein the computing system takes the second action if the determination is that the detected rate of increase of the attribute is at or above the threshold rate of increase of the attribute.
8. The method of claim 1 ,
wherein detecting the pattern of change of the attribute comprises detecting a rate of decrease of the attribute,
wherein making the determination comprises making a determination of whether the detected rate of decrease of the attribute is (i) below a threshold rate of decrease of the attribute or rather (ii) at or above the threshold rate of decrease of the attribute,
wherein the computing system takes the first action if the determination is that the detected rate of decrease of the attribute is below the threshold rate of decrease of the attribute, and
wherein the computing system takes the second action if the determination is that the detected rate of decrease of the attribute is at or above the threshold rate of decrease of the attribute.
9. The method of claim 1 ,
wherein detecting the pattern of change of the attribute comprises detecting a particular change of the attribute, and
wherein making the determination comprises making a determination of whether the particular change corresponds to an increase of the attribute or rather to a decrease of the attribute.
10. The method of claim 9 ,
wherein the computing system takes the first action if the determination is that the particular change corresponds to the increase of the attribute, and
wherein the computing system takes the second action if the determination is that the particular change corresponds to the decrease of the attribute.
11. The method of claim 9 ,
wherein the computing system takes the first action if the determination is that the particular change corresponds to the decrease of the attribute, and
wherein the computing system takes the second action if the determination is that the particular change corresponds to the increase of the attribute.
12. The method of claim 1 ,
wherein detecting the pattern of change of the attribute comprises detecting that the attribute increased,
wherein making the determination comprises making a determination of whether the attribute increased beyond an upper limit of the attribute,
wherein the computing system takes the first action if the determination is that the attribute did not increase beyond the upper limit of the attribute, and
wherein the computing system takes the second action if the determination is that the attribute increased beyond the upper limit of the attribute.
13. The method of claim 1 ,
wherein detecting the pattern of change of the attribute comprises detecting that the attribute decreased,
wherein making the determination comprises making a determination of whether the attribute decreased beyond a lower limit of the attribute,
wherein the computing system takes the first action if the determination is that the attribute did not decrease beyond the lower limit of the attribute, and
wherein the computing system takes the second action if the determination is that the attribute decreased beyond the lower limit of the attribute.
14. The method of claim 1 ,
wherein detecting the pattern of change of the attribute comprises detecting the attribute changed by a particular increment,
wherein making the determination comprises making a determination of whether the particular increment matches an expected increment of change for the attribute,
wherein the computing system takes the first action if the determination is that the particular increment matches the expected increment, and
wherein the computing system takes the second action if the determination is that the particular increment does not match the expected increment.
15. The method of claim 1 ,
wherein taking the first action comprises one or more of the following:
(i) storing an indication that the sport-related information is accurate,
(ii) causing presentation of the sport-related information,
(iii) outputting a report related to the sport-related information being accurate,
(iv) updating a user-account based on the sport-related information, and
(v) causing presentation of supplemental content based on the sport-related information, and
wherein taking the second action comprises one or more of the following:
(i) storing an indication that the sport-related information is inaccurate,
(ii) deleting the sport-related information,
(iii) outputting a report related to the sport-related information being inaccurate, and
(iv) triggering a machine-driven procedure to again extract the sport-related information from the time sequence of digital video frames that represent the sport event.
16. A computing system comprising:
a processing unit;
non-transitory data storage; and
program instructions stored in the non-transitory data storage and executable by the processing unit to carry out operations for determining accuracy of sport-related information extracted from a time sequence of digital video frames that represent a sport event, wherein the extracted sport-related information includes an attribute that changes over the time sequence, the operations comprising:
detecting, based on the extracted sport-related information, a pattern of change of the attribute over the time sequence, wherein the computing system has access to mapping data that indicates, for each of a plurality of sport events, an expected pattern of change of the attribute,
identifying the sport event represented by the digital video frames,
determining the expected pattern of change of the attribute indicated by the mapping data for the identified sport event,
making a determination of whether the detected pattern of change of the attribute over the time sequence is the expected pattern of change of the attribute indicated by the mapping data for the identified sport event,
if the determination is that the detected pattern of change of the attribute over the time sequence is the expected pattern of change of the attribute indicated by the mapping data for the identified sport event, then, responsive to making the determination, taking a first action that corresponds to the sport-related information being accurate, and
if the determination is that the detected pattern of change of the attribute over the time sequence is not the expected pattern of change of the attribute indicated by the mapping data for the identified sport event, then, responsive to making the determination, taking a second action that corresponds to the sport-related information being inaccurate.
17. The computing system of claim 16 , wherein the attribute is (i) a score associated with the sport event or (ii) a time depicted by the digital video frames in accordance with the sport event.
18. A non-transitory computer readable medium having stored thereon instructions executable by a processing unit to cause a computing system to perform operations for determining accuracy of sport-related information extracted from a time sequence of digital video frames that represent a sport event, wherein the extracted sport-related information includes an attribute that changes over the time sequence, the operations comprising:
detecting, based on the extracted sport-related information, a pattern of change of the attribute over the time sequence, wherein the computing system has access to mapping data that indicates, for each of a plurality of sport events, an expected pattern of change of the attribute;
identifying the sport event represented by the digital video frames;
determining the expected pattern of change of the attribute indicated by the mapping data for the identified sport event;
making a determination of whether the detected pattern of change of the attribute over the time sequence is the expected pattern of change of the attribute indicated by the mapping data for the identified sport event;
if the determination is that the detected pattern of change of the attribute over the time sequence is the expected pattern of change of the attribute indicated by the mapping data for the identified sport event, then, responsive to making the determination, taking a first action that corresponds to the sport-related information being accurate; and
if the determination is that the detected pattern of change of the attribute over the time sequence is not the expected pattern of change of the attribute indicated by the mapping data for the identified sport event, then, responsive to making the determination, taking a second action that corresponds to the sport-related information being inaccurate.