CN107203646A - A kind of intelligent social sharing method and device - Google Patents

Abstract

The present application discloses a method and device for intelligent social sharing, wherein the method includes: when it is necessary to generate a VR picture for a shared picture posted by a user, obtain the information associated with the user's current location from a preset geographical picture database system panorama picture; extract feature information from the shared picture and the panorama picture; according to the feature information, filter out a group of panorama pictures from the panorama picture that have a predetermined degree of association with the shared picture, and filter the panorama picture 360-degree stitching is performed on the panorama picture and the shared picture, and the result of the stitching is set as a virtual reality VR picture of the shared picture. By adopting the present invention, an empathetic sharing experience can be provided, and the implementation cost is low and easy to implement.

Description

A method and device for intelligent social sharing

technical field

The invention relates to computer application technology, in particular to an intelligent social sharing method and device.

Background technique

Social software is one of the most frequently used applications in daily life. Many people will share their experiences and status in social software. More interesting and vivid status sharing will add highlights to social software. With the maturity of virtual reality, augmented reality, artificial intelligence and other technologies, it provides richer application scenarios and more vivid user experience for social software.

At present, a solution is proposed that allows people to go anywhere instantly, allowing people to travel around the world without leaving their homes. This solution combines electronic maps, sensors and virtual reality technology through the network, so that users can send back images and sounds to the user's VR glasses through other users' sensors, so that users can use the "separation" or "possession" method, Appearing anywhere in reality at any time satisfies people's travel needs to a large extent, so that people can obtain similar effects to real people's travel without having to travel by car.

In the above solution, other users need to send sensor data back to watch with VR glasses. Therefore, additional sensor devices need to be added. Moreover, you need to use a panoramic camera when taking pictures, and the panoramic camera is a fisheye camera. Photos taken with a fisheye camera can only be viewed with a VR device, and there will be distortion when viewed with other cameras. Therefore, it is necessary to change the user's usage habits, so that The implementation cost of the above scheme is relatively high, and it is not easy to popularize and apply.

Contents of the invention

In view of this, the main purpose of the present invention is to provide an intelligent social sharing method and device, which can provide an empathetic sharing experience, and the implementation cost is low and easy to implement.

In order to achieve the above object, the technical scheme proposed by the present invention is:

A smart social sharing method comprising:

When a VR picture needs to be generated for a shared picture released by the user, the panoramic picture associated with the current location of the user is obtained from the preset geographic picture database system;

Extract characteristic information from the shared picture and the panoramic picture; according to the characteristic information, filter out a group of panoramic pictures with a predetermined degree of association with the shared picture from the panoramic picture, and filter out the panoramic pictures The picture and the shared picture are spliced in 360 degrees, and the spliced one is set as a virtual reality VR picture of the shared picture.

An intelligent social sharing device, comprising:

A panorama image acquisition unit, used to generate a VR image from a preset geographical image database system associated with the user's current location when it is necessary to generate a VR image for a shared image released by the user;

A VR picture generation unit, configured to extract feature information from the shared picture and the panoramic picture; according to the feature information, select a group of panoramic pictures with a predetermined degree of relevance from the panoramic pictures, and filter out Perform 360-degree splicing of the panoramic picture and the shared picture, and set the spliced result as a virtual reality VR picture of the shared picture.

To sum up, the intelligent social sharing method and device proposed by the present invention, based on the picture shared by the user, determines the panoramic picture that matches the shared picture, and then splices the shared picture into a VR picture that can be viewed through a VR device. In this way, VR images can be used to provide an empathetic sharing experience. Moreover, since the present invention relies on other terminal devices to be additionally equipped with sensors, the implementation cost of VR pictures obtained by the present invention is relatively low, easy to implement, and does not require changing the user's usage habits.

Description of drawings

Fig. 1 is the schematic flow chart of the method of the embodiment of the present invention;

Fig. 2 is a schematic diagram of sharing pictures in an embodiment of the present invention;

Fig. 3 is a panoramic picture obtained in an embodiment of the present invention;

Fig. 4 is the panorama two picture that obtains in an embodiment of the present invention;

Figure 5 is an enlarged view of the shared picture in Figure 2;

FIG. 6 is a panorama-360-degree picture selected in an embodiment of the present invention;

FIG. 7 is a 360-degree mosaic picture of a panorama picture and a shared picture selected in an embodiment of the present invention;

Fig. 8 is a 360-degree mosaic image of Fig. 7 after being rendered by VR playback;

Fig. 9 is a schematic diagram of the device structure of the embodiment of the present invention.

detailed description

In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Fig. 1 is the schematic flow chart of the method of the embodiment of the present invention, as shown in Fig. 1, the intelligent social sharing method realized by this embodiment mainly includes:

Step 101, when it is necessary to generate a VR picture for a shared picture posted by a user, obtain a panoramic picture associated with the user's current location from a preset geographic picture database system.

In this step, it is necessary to obtain the panorama picture corresponding to the shared picture from the preset geographical picture database system, so as to filter out the panorama picture matching the shared picture thereafter for generating the corresponding VR picture. Since the corresponding panoramic pictures are obtained from the preset geographical picture database system, other terminals in the corresponding positions do not need to send back pictures through sensors, so the implementation cost of this embodiment can be effectively reduced, and there is no need to change the user's Terminal usage habits.

Preferably, the geographic picture database system can be an existing network geographic information system, namely WebGIS system.

Preferably, in order to filter out panoramic pictures with a high degree of matching with the user's photo in the subsequent process, all panoramic pictures associated with the current location of the user may be obtained here.

In practical applications, the following methods can be used to determine the current location of the user:

The current location of the user is determined according to the GPS information of the terminal device used by the user, the physical address of the WIFI node used or the identification number of the serving base station.

The above three specific implementations for determining the user location are within the grasp of those skilled in the art, and will not be repeated here.

Step 102, extract characteristic information from the shared picture and the panoramic picture; according to the characteristic information, filter out a group of panoramic pictures with a predetermined degree of association with the shared picture from the panoramic picture, and filter the selected 360-degree stitching is performed on the panorama picture and the shared picture, and the stitched result is set as a virtual reality VR picture of the shared picture.

Preferably, the feature information includes: real objects in the picture, text in the picture, location information in the text, location information of the photo and color information of the picture, and the real object includes people, animals and scenery.

In actual implementation, the combination of the existing HOG, Harr and SIFT three commonly used feature extraction algorithms can be used to extract the above feature information from the shared pictures.

Furthermore, in order to enable viewers of VR pictures to better obtain an empathetic sharing experience, further rendering processing can be performed on the 360 stitched pictures obtained after stitching in this step, highlighting the focus and mood of the sharer, specifically However, this can be achieved in the following ways:

Extract the user's focus and mood from the shared pictures, and perform VR playback and rendering on the splicing result according to the focus and mood.

Preferably, in the above method, the following strategies can be used to extract the user's focus and mood:

When the proportion of the area of the real object in the shared picture in the picture is greater than a preset threshold, the real object is determined as the focus of the user.

The threshold described here is used to determine the user's attention point. When the area occupied by the real object in the picture exceeds a certain threshold, it can be identified as the user's attention point. The threshold can be set by those skilled in the art according to actual needs. Appropriate value .

When the shared picture includes a person, the user's mood is extracted according to the expression of the person.

When the shared picture includes text with emotional color, the user's mood is extracted from the text.

The above is only an optional strategy for extracting user concerns and moods. In practical applications, other factors can also be considered to accurately determine user concerns and moods. For example, specific types and locations of objects in pictures can be considered. , such as giving priority to the object in the center as the focus point, giving priority to the person occupying a larger proportion in the picture as the focus point, etc. The specific strategy is not limited to the above, and can be set by those skilled in the art according to actual needs, so I won’t go into details here. .

Preferably, the predetermined degree of association can be set to have the highest degree of association with the shared picture.

Preferably, in the above method, the following method can be used to perform VR playback rendering on the splicing result:

On the picture corresponding to the splicing result, highlight and render the user's focus, add corresponding mood text, and/or set background music matching the mood for the VR playback of the splicing result.

The intelligent social sharing method realized by the present invention will be further elaborated below through the realization process of the present invention in a specific scenario:

1) The user takes a picture with a mobile phone, and distributes text to the picture for sharing, as shown in Figure 2.

2) The GPS of the mobile phone obtains the latitude and longitude, and locates the location of the photo as Mount Sanqingshan.

3) Obtain the panoramic picture of Sanqingshan Mountain through WebGIS, and the panoramic picture includes all the panoramas of Sanqingshan Mountain. Fig. 3 is the panorama picture obtained, from which the character feature value on the picture is extracted, and the picture feature value is set as "famous mountain in the world". Fig. 4 is the second panorama picture obtained, from which the feature value of the shape on the picture is extracted, and the feature value is set as "the shape is a tower". Extract the features of the shared picture (Figure 3), as shown in the enlarged picture of Figure 2 and Figure 5, it can be analyzed that there are signs of "Famous Mountains in the World" and "Sanqing Fudi" in the pictures shared by users, and this can be used as the semantic meaning of the picture. Feature extraction items.

4) Filter the WebGIS photos according to the user's photos: the feature extraction items of the user's shared pictures include the logos of "Famous Mountains in the World" and "Sanqing Fudi". The feature value of the first panorama picture is "Famous Mountains in the World". , the panorama-1 picture has a higher correlation with the eigenvalues of the user-shared pictures, so the panorama-1 picture is selected as the material for 360-degree picture stitching, as shown in Figure 6.

5) Match and fuse the user's shared picture with the WebGIS picture to form a new 360-degree panoramic photo, as shown in Figure 7.

6) Analyze the user's shared photos to determine the user's focus and mood. From the text of the shared pictures, positive expressions such as "good" and "good" can be obtained, and it can be judged that the user is in a good mood. In the photo, the proportion of the portrait is relatively large and it is located in the center of the photo. It can be determined that the user's focus is on the portrait.

7) According to the user's focus and mood, perform VR rendering on the splicing result, intelligently highlight the focus and match the mood text, so that the viewer can quickly grasp the key points shared by the sharer, as shown in Figure 8.

It should be noted that the pictures in Figure 2-8 above are color pictures, and in order to meet the requirements of the accompanying drawings, they are converted into black and white pictures. Color matching.

In practical application, the present invention can be implemented in the sharer's terminal device, for example, when the sharer publishes the shared picture, the present invention can be used to generate the corresponding VR picture; it can also be implemented in the server of the social sharing software, for example When a viewer who shares a picture requests the server to view a corresponding VR picture, the server triggers the execution of the present invention according to the request, thereby generating a corresponding VR picture and sending it to the viewer. Specifically, a person skilled in the art may select an appropriate implementation entity according to actual needs.

Fig. 9 is a schematic structural diagram of an intelligent social sharing device corresponding to the above method. As shown in Fig. 9, the device includes:

The panorama picture acquiring unit is used for generating a VR picture for a shared picture released by the user, from a preset geographic picture database system to obtain a panorama picture associated with the user's current location.

A VR picture generation unit, configured to extract feature information from the shared picture and the panoramic picture; according to the feature information, select a group of panoramic pictures with a predetermined degree of relevance from the panoramic pictures, and filter out Perform 360-degree splicing of the panoramic picture and the shared picture, and set the spliced result as a virtual reality VR picture of the shared picture.

Preferably, the panoramic picture acquisition unit is configured to determine the current location of the user according to the GPS information of the terminal device used by the user, the physical address of the WIFI node or the identification number of the serving base station.

Preferably, the feature information includes: real objects in the picture, text in the picture, location information in the text, location information of the photo and color information of the picture, and the real object includes people, animals and scenery.

Preferably, the VR picture generation unit is further configured to extract the user's focus and mood from the shared picture, and perform VR playback and rendering on the splicing result according to the focus and mood.

Preferably, the VR picture generating unit is configured to determine the real object as the user's attention point when the area ratio of the real object in the shared picture in the picture is greater than a preset threshold; when the shared picture When a character is included, the user's mood is extracted according to the expression of the character; when the shared picture includes text with emotional color, the user's mood is extracted from the text.

Preferably, the VR picture generating unit is configured to highlight and render the user's focus on the picture corresponding to the splicing result, add corresponding mood text, and/or set the VR playback of the splicing result to be consistent with the splicing result. background music to match the mood.

Preferably, the geographic picture database system is a WebGIS system.

Preferably, the panorama picture acquisition unit is further configured to acquire all panorama pictures associated with the current location of the user.

Preferably, the predetermined degree of association is the highest degree of association with the shared picture.

To sum up, the above are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (18)

1. A method for intelligent social sharing, characterized in that, comprising: When a VR picture needs to be generated for a shared picture released by the user, the panoramic picture associated with the current location of the user is obtained from the preset geographic picture database system; Extract characteristic information from the shared picture and the panoramic picture; according to the characteristic information, filter out a group of panoramic pictures with a predetermined degree of association with the shared picture from the panoramic picture, and filter out the panoramic pictures A 360-degree splicing is performed on the picture and the shared picture, and the spliced result is set as a virtual reality VR picture of the shared picture. 2. The method according to claim 1, wherein the method for determining the current location of the user comprises: The current location of the user is determined according to the GPS information of the terminal device used by the user, the physical address of the WIFI node, or the identification number of the serving base station. 3. The method according to claim 1, wherein the feature information includes: the real object in the picture, the text in the picture, the location information in the text, the location information of the photo, and the color information of the picture, and the real object Including people, animals and scenery. 4. The method according to claim 1, wherein the method further comprises: Extract the user's focus and mood from the shared pictures, and perform VR playback and rendering on the splicing result according to the focus and mood. 5. method according to claim 4, is characterized in that, described extraction user's focus and mood comprise: When the area ratio of the real object in the shared picture in the picture is greater than a preset threshold, the real object is determined as the user's attention point; When a person is included in the shared picture, extracting the mood of the user according to the expression of the person; When the shared picture includes text with emotional color, the user's mood is extracted from the text. 6. The method according to claim 4, wherein the VR playback rendering includes: highlighting and rendering the user's attention points on the picture corresponding to the splicing result, adding corresponding mood text and/or providing Set the background music matching the mood for the VR playback of the splicing result. 7. The method according to claim 1, characterized in that, the geographic picture database system is a WebGIS system. 8. The method according to claim 1, wherein said acquiring comprises: Obtain all panoramic pictures associated with the current location of the user. 9. The method according to claim 1, wherein the predetermined degree of association is the highest degree of association with the shared picture. 10. An intelligent social sharing device, characterized in that it comprises: A panorama image acquisition unit, used to generate a VR image from a preset geographical image database system associated with the user's current location when it is necessary to generate a VR image for a shared image released by the user; A VR picture generation unit, configured to extract feature information from the shared picture and the panoramic picture; according to the feature information, select a group of panoramic pictures with a predetermined degree of relevance from the panoramic pictures, and filter out The panoramic picture and the shared picture are 360-degree spliced, and the spliced is set as the virtual reality VR picture of the shared picture. 11. The device according to claim 10, wherein the panorama image acquisition unit is configured to determine the panoramic image according to the GPS information of the terminal equipment used by the user, the physical address of the WIFI node, or the identification number of the serving base station. The user's current location. 12. The device according to claim 10, wherein the feature information includes: the real object in the picture, the text in the picture, the location information in the text, the location information of the photo, and the color information of the picture, and the real object Including people, animals and scenery. 13. The device according to claim 10, wherein the VR picture generation unit is further configured to extract the user's focus and mood from the shared picture, and to make the splicing according to the focus and mood The result is rendered for VR playback. 14. The device according to claim 13, wherein the VR picture generating unit is configured to determine the real object in the shared picture as having an area ratio greater than a preset threshold value in the picture. The user's focus; when the shared picture includes a character, extract the user's mood according to the expression of the character; when the shared picture includes text with emotional color, extract the user's mood from the text Feeling. 15. The device according to claim 13, wherein the VR picture generating unit is configured to highlight and render the user's focus on the picture corresponding to the splicing result, add corresponding mood text and/or Background music matching the mood is set for the VR playback of the splicing result. 16. The device according to claim 10, characterized in that the geographic picture database system is a WebGIS system. 17. The device according to claim 10, wherein the panorama picture acquisition unit is configured to acquire all panorama pictures associated with the current location of the user. 18. The device according to claim 10, wherein the predetermined degree of association is the highest degree of association with the shared picture.