System for providing a metaverse-based virtualized image and method therefor
The present disclosure provides a method of providing a metaverse-based virtualized image, the method including capturing an image in real time via an IP camera, storing the captured image, recognizing at least one face and object from the captured image, virtualizing the recognized at least one face and object, and outputting a virtualized image information including a virtualized character and a virtual object.
1. A method of providing a metaverse-based virtualized image, the method comprising:
capturing an image in real time via an internet protocol (IP) camera;
storing the captured image;
recognizing a face and an object from the captured image;
generating a virtual character and a virtual object corresponding to a virtualization of the recognized face and the object, respectively, based on the captured image; and
outputting a virtualized image information including the virtualized character and the virtual object,
wherein recognizing the face and the object from the captured image comprises:
requesting an analysis of the captured image by transmitting frame and sound information of the captured image to a cloud server at a preset interval period; and
detecting a change of the recognized face and object,
wherein a length of the preset interval period is set to be shorter in a case in which motion of at least one of the recognized face or the recognized object is detected compared to a longer length in a case in which no motion is detected.
2. The method of claim 1 , wherein the face and the object of the captured image are recognized by the cloud server based on an artificial intelligence (AI) algorithm, and wherein the AI algorithm includes at least one of object detection, color detection, face identification, object tracking, action detection, or sound analysis.
3. The method of claim 2 , wherein the virtual character and virtual object are generated by 3D-modeling the recognized face and 3D-modeling the recognized object.
4. The method of claim 3 , wherein locations of the virtual character and the virtual object in the virtualized image information respectively correspond to locations of the recognized face and recognized object in the captured image.
5. The method of claim 4 , further comprising saving the virtualized image information to the cloud server, wherein the virtualized image information is shared with a terminal by the cloud server.
6. The method of claim 5 , wherein a metaverse image is generated and output by the terminal based on the virtualized image information and a virtualized image stored in the terminal.
7. The method of claim 6 , further comprising:
detecting a preset situation; and
providing a notification based on detection of the preset situation.
8. The method of claim 7 , wherein based on detection of the preset situation, an original copy of the captured image is saved to the cloud server.
9. The method of claim 8 , wherein the preset situation includes:
wherein at least one of an unregistered character or a virtual object is recognized from the virtualized image information,
wherein a registered character is recognized from the virtualized image information, or
wherein the registered character does not move during a preset time based on the virtualized image information.
10. The method of claim 9 , further comprising:
receiving a second virtualized image information including a second virtual object from the cloud server;
outputting the second virtualized image information;
receiving a signal of selecting the second virtual object;
searching for an object related or similar to the second virtual object; and
outputting the object resulting from the search.
11. The method of claim 1 , wherein the captured image is stored in a memory of the IP camera.
12. A system for providing a metaverse-based virtualized image, the system comprising:
an IP camera configured to capture an image;
a cloud server configured to: receive frame and sound information of the captured image, recognize a face and an object from the captured image based on an AI algorithm, generate a virtual character and a virtual object corresponding to a virtualization of the recognized face and the object, respectively, based on the captured image; and
a terminal configured to receive virtualized image information from the cloud server and output the received virtualized image information,
wherein the terminal is further configured to:
request an analysis of the captured image by transmitting frame and sound information of the captured image to a cloud server at a preset interval period; and
detect a change of the recognized face and object,
wherein a length of the preset interval period is set to be shorter in a case in which motion of at least one of the recognized face or the recognized object is detected compared to a longer length in a case in which no motion is detected.