CN-121982173-A - Shadow play digital human interaction method, system, equipment and medium

CN121982173ACN 121982173 ACN121982173 ACN 121982173ACN-121982173-A

Abstract

The application discloses a shadow-play digital human interaction method, a system, equipment and a medium, wherein the method comprises the steps of firstly acquiring instruction information input by a user, then inquiring a corresponding shadow-play action sequence from a motion mapping table according to the instruction information, wherein the action sequence comprises an initial action section and a subsequent action section, each action section comprises a plurality of continuous action frames, taking a target reference joint of a first frame in the action sequence as a base line, after determining a target reference joint coordinate difference of a start frame and a stop frame of the action sections, carrying out translation alignment on joint coordinates of the action sections according to the base line and the coordinate difference, obtaining a corrected action sequence according to the joint coordinates, and rendering the action sequence, thereby obtaining a dynamic deduction picture. According to the embodiment of the application, through the design of instruction-driven motion matching and motion segment translation alignment based on the reference joint, the interaction between the shadow-play digital person and the user and the consistent and stable presentation of motion span switching are realized, and the viewing experience is effectively improved.

Inventors

QIN CHUANBO
LI YIXUAN
GUO QI
Kuang Yaoxi
YAO YONGKANG
ZENG JUNYING

Assignees

五邑大学

Dates

Publication Date: 20260505
Application Date: 20251212

Claims (10)

1. The shadow play digital human interaction method is characterized by comprising the following steps of: acquiring instruction information input by a user; inquiring a corresponding shadow play action sequence from a preset action mapping table according to the instruction information, wherein the shadow play action sequence comprises an initial action section and at least one subsequent action section, and each action section comprises a plurality of continuous action frames; taking a target reference joint of a first frame in the shadow play action sequence as a base line; Determining a target reference joint coordinate difference of a start-stop frame of the initial action segment and a start-stop frame of a subsequent action segment; According to the baseline and the corresponding target reference joint coordinate difference, carrying out translation alignment on joint coordinates of an action section to obtain joint coordinates after translation alignment; And obtaining a corrected shadow play action sequence according to the joint coordinates after translation alignment, and rendering the corrected shadow play action sequence to obtain a shadow play digital human dynamic deduction picture.
2. The method according to claim 1, wherein the step of constructing the action map comprises: Collecting a human body performance action sequence, wherein the performance action sequence comprises a plurality of performance frames, and each performance frame corresponds to a plurality of skeleton key points of a human body; Screening out core skeleton key points of the shadow-play digital person from the skeleton key points, and carrying out coordinate normalization processing on the core skeleton key points by taking target reference joint coordinates as reference points to obtain processed key point coordinates; obtaining a custom performance action sequence according to the processed key point coordinates; And associating the custom performance action sequence with the corresponding semantic instruction, and then adding the associated custom performance action sequence into an initial mapping table to obtain an action mapping table.
3. The method of claim 2, wherein the core skeletal keypoints comprise at least one of a head keypoint, a left shoulder keypoint, a right shoulder keypoint, a left elbow keypoint, a right elbow keypoint, a left wrist keypoint, a right wrist keypoint, a left hip keypoint, a right hip keypoint, a left knee keypoint, a right knee keypoint, a left ankle keypoint, a right ankle keypoint.
4. The method of claim 1, wherein prior to rendering the modified shadow play motion sequence, the method further comprises performing joint interpolation operations on the modified shadow play motion sequence to obtain a shadow play motion sequence with continuous and smooth motion.
5. The method of claim 1, wherein rendering the modified shadow play action sequence comprises: loading the component materials of the shadow digital man; Mapping joint coordinates of an action frame in the corrected shadow play action sequence to a corresponding shadow play component; and correcting the position of the middle point of the shoulder, the offset of the shoulder and the height of the elbow of each shadow-play part, and rendering each shadow-play part frame by frame based on a preset frame rate.
6. The method of claim 1, wherein the querying the corresponding shadow play action sequence from the preset action mapping table according to the instruction information includes: extracting keywords and carrying out semantic analysis on the instruction information to obtain a corresponding text instruction; and inquiring a corresponding shadow play action sequence from a preset action mapping table according to the text instruction.
7. The method according to claim 1, wherein after acquiring instruction information input by a user, the method further comprises: Generating adaptive response information according to the instruction information; and displaying the response information.
8. The shadow play digital human interaction system is characterized by comprising: The input module is used for acquiring instruction information input by a user; the motion inquiry module is used for inquiring a corresponding shadow play motion sequence from a preset motion mapping table according to the instruction information, wherein the shadow play motion sequence comprises an initial motion segment and at least one subsequent motion segment, and each motion segment comprises a plurality of continuous motion frames; The joint correction module is used for taking a target reference joint of a first frame in the shadow play action sequence as a base line, determining a target reference joint coordinate difference of a start-stop frame of the initial action segment and a start-stop frame of a subsequent action segment, and carrying out translation alignment on joint coordinates of the action segment according to the base line and the corresponding target reference joint coordinate difference to obtain joint coordinates after translation alignment; and the action rendering module is used for obtaining a corrected shadow-play action sequence according to the joint coordinates after translation alignment, and rendering the corrected shadow-play action sequence to obtain a shadow-play digital human dynamic deduction picture.
9. An electronic device comprising a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program when executed by the processor implements the method of any of claims 1 to 7.
10. A computer readable storage medium, in which a processor executable program is stored, characterized in that the processor executable program is for performing the method according to any of claims 1 to 7 when being executed by a processor.

Description

Shadow play digital human interaction method, system, equipment and medium Technical Field The embodiments of the present application relate to but are not limited to the field of shadow mask design, in particular to a shadow play digital human interaction method systems, devices, and media. Background The traditional shadow play is taken as an important non-material cultural heritage in China, and the inheritance and the propagation of the shadow play are limited by the dependence on manual operation and the constraint of performance sites for a long time, so that the shadow play is difficult to realize large-scale popularization. Along with the gradual application of the digital technology in the cultural field, the digital display of shadow play becomes an important non-genetic direction of protection and transmission, but the prior proposal still has a plurality of defects. Meanwhile, the existing digital shadow is generated and lacks the effective position calibration mechanism, cause the easy problem of position deviation, linking incoherence of movements while switching over the section, and these schemes are mostly insufficient to consider the unique artistic performance logic of shadow, difficult to present its peculiar artistic style, have seriously influenced and watched the experience, have restricted the shadow to digitize and land and spread widely in the actual scene. Disclosure of Invention The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims. The embodiment of the application provides a shadow-play digital person interaction method, a shadow-play digital person interaction system, shadow-play digital person interaction equipment and a shadow-play digital person interaction medium, which realize the interaction between shadow-play digital person and user and the consistent and stable presentation of action span switching, and effectively promote the viewing experience. In a first aspect, an embodiment of the present application provides a shadow-play digital human interaction method, including obtaining instruction information input by a user; inquiring a corresponding shadow-play action sequence from a preset action mapping table according to the instruction information, wherein the shadow-play action sequence comprises an initial action section and at least one subsequent action section, each action section comprises a plurality of continuous action frames, a target reference joint of a first frame in the shadow-play action sequence is used as a base line, a target reference joint coordinate difference of a start-stop frame of the initial action section and a target reference joint coordinate difference of a subsequent action section are determined, the joint coordinates of the action sections are subjected to translation alignment according to the base line and the corresponding target reference joint coordinate difference to obtain joint coordinates after translation alignment, a modified shadow-play action sequence is obtained according to the joint coordinates after translation alignment, and the modified shadow-play action sequence is rendered to obtain a shadow-play digital human dynamic deduction picture. In combination with the first aspect, in an embodiment of the present application, the step of constructing the action mapping table includes collecting a human performance action sequence, where the performance action sequence includes a plurality of performance frames, each performance frame corresponds to a plurality of skeleton key points of a human body, screening core skeleton key points of the shadow-play digital person from the plurality of skeleton key points, performing coordinate normalization processing on the core skeleton key points with target reference joint coordinates as reference points to obtain processed key point coordinates, obtaining a custom performance action sequence according to the processed key point coordinates, and adding the custom performance action sequence and corresponding semantic instructions to an initial mapping table to obtain the action mapping table. In an embodiment of the present application, the core skeleton key points include at least one of a head key point, a left shoulder key point, a right shoulder key point, a left elbow key point, a right elbow key point, a left wrist key point, a right wrist key point, a left hip key point, a right hip key point, a left knee key point, a right knee key point, a left ankle key point, and a right ankle key point. In an embodiment of the present application, before the corrected shadow play action sequence is rendered, the method further includes performing joint interpolation operation on the corrected shadow play action sequence to obtain a shadow play action sequence with continuous and smooth actions. In an embodiment of the application, the rendering of the modified shadow play action sequence comprises loa