Automated Response in Complex | Artificial Intelligence Situations - iOmniscient

iOmniscient

30 Jul 201502:06

Summary

TLDRIan's video analytics system excels in complex environments, identifying and tracking a dropped bag obscured by crowds. The technology can rewind to the bag's introduction, capture the culprit's face, and add it to a watch list. It then shares this information across networked cameras, enabling real-time tracking and alerting nearby officers via smartphone. The innovative process, from object detection to automated response, is patented, showcasing advanced capabilities in uncontrolled settings.

Takeaways

📹 Ian's specialty is in providing video analytics for complex scenes.
👜 The system can detect and track objects, even when they are obscured by people or other objects.
🔍 The system has the capability to 'jump to a vent,' which allows it to trace back to the moment an object was first introduced into the scene.
👤 It can identify and enroll faces into a watch list, regardless of how crowded or uncontrolled the environment is.
📡 The face recognition feature can be used to search for individuals across various cameras within a network.
👮‍♂️ The system can locate the nearest police officer and send information about a person of interest to their smartphone.
🤔 The police can request additional information about the person's identity or actions through the system.
🔗 The system can send a 'jump to event' video, providing context or evidence of the person's actions.
🚔 The process of automated response, from detection to police notification, is patented by Ian.
🏆 Ian's technology is recognized for its innovative approach to video analytics in crowded and dynamic environments.
🛡️ The entire process, from detecting an object to identifying and tracking individuals, is protected by patents.

Q & A

What is Ian's specialty in video analytics?
-Ian's specialty is providing multiple video analytics in complex scenes, especially where objects are significantly obscured.
How does the system identify a dropped bag in a crowded scene?
-The system is capable of finding a bag even when it is obscured by people walking in front of it and hardly visible.
What is the 'jump to vent' feature used for?
-The 'jump to vent' feature allows the system to go back to the moment when the bag was first brought into the scene, regardless of how long ago it was.
Can the system identify who left the bag even if the person's face is not visible?
-Yes, the system can identify the person who left the bag by pulling out his face and enrolling him in the watch list.
How is the person's face transmitted to other cameras within the network?
-The face is transmitted to all other cameras within the network, allowing for a search across various cameras for the person.
What additional information can be pulled out about the person when they are seen again?
-When the person is seen again, the system can pull out whatever additional information it has about the person.
How does the system track the person across different cameras?
-The system tracks the person on different cameras where he has been seen, using the information it has on him.
What happens when the person is tracked to the last camera shown in red?
-The system finds the nearest policeman and sends the information to him on his smartphone.
How does the system assist the policeman in identifying the person?
-The system can send the policeman as much information as it has on the person, including a 'jump to event' video.
What is the significance of the entire process being patented?
-The patent ensures that the ability to find a bag in a crowded scene, the 'jump to event' feature, face recognition in an uncontrolled environment, and the automated response process are protected and unique to Ian.
What is the role of the smartphone in the process described in the script?
-The smartphone is used by the nearest policeman to receive information about the person of interest and to view the 'jump to event' video.

Outlines

00:00

👀 Advanced Video Analytics for Security

The script introduces Ian's specialty in advanced video analytics, showcasing its ability to detect and track objects, such as a dropped bag, even in complex and crowded scenes. The system can identify the person who left the bag, despite it being obscured, and has the capability to 'jump' back to the moment the bag was first seen. It can also enroll the person's face into a watch list and transmit this information across a network of cameras. The technology includes face recognition in uncontrolled environments and the ability to track individuals across different cameras, ultimately alerting the nearest police officer via smartphone with the suspect's information and the event video. The entire automated response process is patented by ient.

Mindmap

Keywords

💡Video Analytics

Video analytics refers to the use of computer algorithms to analyze and interpret video content for various applications such as security, surveillance, and business intelligence. In the context of the video, it is the core technology enabling the recognition and tracking of objects and people in complex scenes, exemplified by the detection of an abandoned bag.

💡Object Obscured

When an object is obscured, it is blocked from view or made difficult to see due to other objects or people in the way. The video script highlights the system's ability to identify and track objects, such as a bag, even when they are significantly obscured by people or other elements in the scene.

💡Jump to Event

The 'Jump to Event' feature allows the system to quickly navigate to a specific moment in the video where an event of interest occurred. In the script, this feature is used to identify the person who left the bag by going back to the moment when the bag was first introduced into the scene.

💡Watch List

A watch list in surveillance systems is a database of individuals or objects that are of particular interest and require monitoring. The video describes enrolling the person who left the bag into a watch list, which enables the system to track and identify him across different cameras.

💡Face Recognition

Face recognition is a biometric technology that identifies or verifies the identity of a person from their facial features. The video script mentions the system's capability to perform face recognition in a crowded and uncontrolled environment, which is crucial for identifying the person who left the bag.

💡Camera Network

A camera network is a system of interconnected cameras that work together to provide comprehensive surveillance coverage. The script describes how the face of the person who left the bag is transmitted to all other cameras within the network, allowing for extensive monitoring.

💡Tracking

Tracking in the context of video analytics involves following the movement of an object or person across different camera feeds. The video script illustrates the system's ability to track the individual who left the bag across various cameras in the network.

💡Nearest Policeman

The term 'nearest policeman' refers to the law enforcement officer who is closest to the location of an incident. The video script explains how the system can identify the nearest policeman and send him information about the person of interest, facilitating a quick response.

💡Smartphone

A smartphone is a handheld device that combines the functions of a mobile phone and a computer. In the script, the smartphone is used by the policeman to receive information about the person of interest, demonstrating the integration of technology in law enforcement operations.

💡Automated Response

Automated response refers to the system's ability to perform actions without human intervention. The video script describes an automated process that includes identifying the abandoned bag, tracking the person, and alerting the nearest policeman, all without direct human control.

💡Patented

A patent is a form of intellectual property that gives its owner the right to exclude others from making, using, or selling an invention for a limited period. The script mentions that the entire process described, including finding an obscured object, jump to event, and face recognition, is patented by the company ient.

Highlights

Ian's specialty in providing multiple video analytics in complex scenes.

System's ability to detect a dropped and abandoned bag even when obscured by people.

Capability to handle situations with significantly obscured objects.

Jump to vent feature to trace back the moment an object was introduced.

System's capacity to identify who left the bag regardless of the time elapsed.

Face enrollment in the watch list for further tracking.

Transmission of the person's face to other cameras within the network.

Search for the person across various cameras in the network.

Extraction of additional information about the person from different camera views.

Tracking the person across different cameras to monitor his movements.

Identification of the last camera where the person was seen, marked in red.

Finding the nearest policeman and sending the person's information to his smartphone.

Ability to provide detailed information about the person to the police.

Sending the 'jump to event' video to explain the person's actions.

The entire automated response process is patented.

Patented technology includes finding obscured objects, jump to event, face recognition, and automated response.

The system operates in uncontrolled environments with high accuracy.