Information processing apparatus, information processing method, and information processing program
AssigneeNEC Corporation
InventorRyoma Oami
Life of this patent
Estimating the term…
Abstract
To efficiency search for an object associated with a sensed event, an information processing apparatus includes a sensor that analyzes a captured video and senses whether a predetermined event has occurred, a determining unit that determines a type of an object to be used as query information based on a type of the event in response to sensing of the event occurrence, and a generator that detects the object of the determined type from the video and generates the query information based on the detected object.
Claims15
An information processing system comprising:
- one or more memories storing:
- time difference information designating a time period of video data at least one before and after luggage abandonment has occurred as a predetermined event;
- and program instructions;
- and one or more processors configured to execute the program instructions to, each of a plurality of times starting with a first time:
- analyze a captured video and sense occurrence of the predetermined event;
- detect an object relating to the predetermined event from the captured video, using a capturing time which is calculated based on a time of the occurrence of the predetermined event and the time difference information, wherein the time difference information that is used the first time the object is detected is predetermined time information, and the time different information that is used each time the object is detect other than the first time is the time difference information as has been updated a preceding time;
- track the detected object;
- extract a feature of the detected object as query information;
- search the captured video using the query information to identify images in the captured video that include the extracted feature;
- output the identified images;
- learn, from a result of detection and tracking of the object, a time difference between the time of the occurrence of the predetermined event and a time when the feature is extracted;
- and update the time difference information using the learned time difference, such that the time different as has been updated is used in detecting the object a next time.
A method performed by a computer and comprising:
- storing time difference information designating a time period of video data at least one before and after luggage abandonment has occurred as a predetermined event;
- each of a plurality of times starting with a first time:
- analyzing a captured video and sense occurrence of the predetermined event;
- detecting an object relating to the predetermined event from the captured video, using a capturing time which is calculated based on a time of the occurrence of the predetermined event and the time difference information, wherein the time difference information that is used the first time the object is detected is predetermined time information, and the time different information that is used each time the object is detect other than the first time is the time difference information as has been updated a preceding time;
- tracking the detected object;
- extracting a feature of the detected object as query information;
- searching the captured video using the query information to identify images in the captured video that include the extracted feature;
- outputting the identified images;
- learning, from a result of detection and tracking of the object, a time difference between the time of the occurrence of the predetermined event and a time when the feature is extracted;
- and updating the time difference information using the learned time difference, such that the time different as has been updated is used in detecting the object a next time.
A non-transitory computer readable medium storing a program causing a computer to execute a process comprising:
- storing time difference information designating a time period of video data at least one before and after luggage abandonment has occurred as a predetermined event;
- each of a plurality of times starting with a first time:
- analyzing a captured video and sense occurrence of the predetermined event;
- detecting an object relating to the predetermined event from the captured video, using a capturing time which is calculated based on a time of the occurrence of the predetermined event and the time difference information, wherein the time difference information that is used the first time the object is detected is predetermined time information, and the time different information that is used each time the object is detect other than the first time is the time difference information as has been updated a preceding time;
- tracking the detected object;
- extracting a feature of the detected object as query information;
- searching the captured video using the query information to identify images in the captured video that include the extracted feature;
- outputting the identified images;
- learning, from a result of detection and tracking of the object, a time difference between the time of the occurrence of the predetermined event and a time when the feature is extracted;
- and updating the time difference information using the learned time difference, such that the time different as has been updated is used in detecting the object a next time.
Dependencies are inferred from the claim text.