Method, electronic device, and computer program product for rendering target scene
Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for rendering a target scene. The method includes acquiring from a server rendering data including a point cloud and a color of the target scene. The method further includes determining a pose used for rendering the target scene at user equipment. The method further includes rendering the target scene on the basis of the rendering data and the pose, so as to show the target scene in the pose. In embodiments of the present disclosure, a rendering effect on the target scene can be improved, and the rendering speed is higher.
1. A method for rendering a target scene, comprising:
acquiring from a server rendering data comprising a point cloud and color information of the target scene;
determining a pose used for rendering the target scene at user equipment; and
rendering the target scene on the basis of the rendering data and the pose, so as to show the target scene in the pose;
wherein rendering the target scene comprises:
determining, on the basis of the rendering data and the pose, a probability that light irradiates a point in the point cloud of the target scene and penetrates through the point; and
rendering the target scene in the pose on the basis of the probability.
2. The method according to claim 1 , wherein the rendering data is first rendering data, and the first rendering data is determined by performing downsampling on second rendering data, a resolution of the second rendering data being higher than that of the first rendering data.
3. The method according to claim 2 , wherein rendering the target scene on the basis of the rendering data and the pose comprises:
performing upsampling on the first rendering data on the basis of the pose to determine the first rendering data after upsampling; and
determining the point cloud and the color information of the target scene in the pose using the first rendering data after upsampling.
4. The method according to claim 1 , wherein the rendering data is determined on the basis of volume density features and color features, which are associated with the target scene, of a plurality of images of the target scene in different poses.
5. The method according to claim 4 , further comprising:
acquiring a scene rendering model from the server; and
rendering the target scene using the scene rendering model.
6. The method according to claim 5 , wherein the scene rendering model is trained on the basis of a plurality of images of the target scene photographed from different poses, as well as poses and internal parameters of a camera respectively corresponding to the plurality of images.
7. A method for rendering a target scene, comprising:
acquiring from a server an edge graph that represents an outline of the target scene;
determining a pose used for rendering the target scene at user equipment; and
rendering the target scene on the basis of the edge graph and the pose, so as to show the target scene in the pose;
wherein rendering the target scene comprises:
determining, on the basis of the edge graph and the pose, a probability that light irradiates a point on the outline of the target scene and penetrates through the point; and
determining a point cloud and color information of the target scene in the pose on the basis of the probability.
8. The method according to claim 7 , further comprising:
acquiring a scene rendering model from the server; and
rendering the target scene using the scene rendering model.
9. The method according to claim 7 , wherein the edge graph is generated on the basis of an edge score of a point in the point cloud of the target scene.
10. The method according to claim 9 , wherein the edge score is determined on the basis of a probability indicating that the point is on the outline of the target scene.
11. The method according to claim 10 , wherein the point with the probability greater than a threshold probability is identified on the edge graph.
12. A method for rendering a target scene, comprising:
acquiring from a server an edge graph that represents an outline of the target scene;
determining a pose used for rendering the target scene at user equipment; and
rendering the target scene on the basis of the edge graph and the pose, so as to show the target scene in the pose;
wherein the method further comprises:
acquiring a scene rendering model from the server; and
rendering the target scene using the scene rendering model; and
wherein the scene rendering model is trained on the basis of a plurality of images of the target scene photographed from different poses, poses and internal parameters of a camera respectively corresponding to the plurality of images, and a pre-labeled edge graph.
13. An electronic device, comprising:
a processor; and
a memory coupled to the processor, wherein the memory has instructions stored therein which, when executed by the processor, cause the device to execute actions comprising:
acquiring from a server rendering data comprising a point cloud and color information of a target scene;
determining a pose used for rendering the target scene at user equipment; and
rendering the target scene on the basis of the rendering data and the pose, so as to show the target scene in the pose;
wherein rendering the target scene comprises:
determining, on the basis of the rendering data and the pose, a probability that light irradiates a point in the point cloud of the target scene and penetrates through the point; and
rendering the target scene in the pose on the basis of the probability.
14. The electronic device according to claim 13 , wherein the rendering data is first rendering data, and the first rendering data is determined by performing downsampling on second rendering data, a resolution of the second rendering data being higher than that of the first rendering data.
15. The electronic device according to claim 14 , wherein rendering the target scene on the basis of the rendering data and the pose comprises:
performing upsampling on the first rendering data on the basis of the pose to determine the first rendering data after upsampling; and
determining the point cloud and the color information of the target scene in the pose using the first rendering data after upsampling.
16. The electronic device according to claim 13 , wherein the rendering data is determined on the basis of volume density features and color features, which are associated with the target scene, of a plurality of images of the target scene in different poses.
17. A computer program product tangibly stored on a non-transitory computer-readable medium and comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a device, cause the device to perform the method according to claim 1 .
18. A computer program product tangibly stored on a non-transitory computer-readable medium and comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a device, cause the device to perform the method according to claim 7 .