Method for collecting three-dimensional motion data through multiple channels

Abstract

The invention discloses a method for collecting three-dimensional motion data through multiple channels, and relates to the technical field of three-dimensional motion, and the method comprises the steps of 1, three-dimensional motion data collection, 2, inertia measurement, 3, motion analysis and 4, three-dimensional motion judgment. The motion capture index of each role in the film and television production process and the light source coincidence index of each role in the film and television production process are analyzed, and the motion fluency index of each role in the film and television production process is evaluated through the motion fluency model, so that the vitality and the infection of the roles are improved, the artistic expressive force of film and television works is improved, and the decision-making efficiency is improved. The manufacturing period is shortened, and the cost caused by reworking is reduced; based on the obtained action fluency index of each role in the film and television production process, whether the action is fluent or not is judged, the shot application level is matched, the accuracy of shot language and action performance is improved, the industrial production efficiency is improved, the subjective trial and error time is shortened, and refined allocation of resources is achieved.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional motion technology, and in particular to a method for collecting three-dimensional motion data through multiple channels. Background Art

[0002] In the context of the deep integration of modern technology and the cultural industry, 3D motion data acquisition technology has become a core support for fields such as film and television production. The emergence of optical motion capture systems has driven technological innovation. Passive optical systems use multiple cameras to capture reflective markers, achieving submillimeter-level precision in 3D coordinate measurement. They are widely used to accurately reproduce character movements in film and television special effects production. However, optical systems are significantly affected by ambient lighting, and occlusion of markers can lead to data loss. Furthermore, the high cost and complex layout of the equipment limit its application in outdoor and small-scale projects. Therefore, it is necessary to analyze the methods for multi-channel acquisition of 3D motion data.

[0003] Prior art, such as the invention patent application with publication number CN114301880B, discloses a 3D data transmission method, electronic device, and signaling server. The method includes: establishing a connection with a signaling server and sending a multi-connection request to the signaling server, wherein the multi-connection request includes a capture terminal identifier and negotiation parameters; after the signaling server determines which rendering terminals belong to the same group as the capture terminal, establishing an end-to-end connection with each rendering terminal in the same group through the signaling server; sending a channel establishment request carrying a data transmission channel identifier corresponding to each rendering terminal through the signaling server to establish a P2P data transmission channel with each rendering terminal; and responding to a 3D data capture instruction, collecting 3D data and sending the 3D data to the corresponding rendering terminal via the established P2P data transmission channel with each rendering terminal. This solves the problem that the end-to-end connection in the data transmission solution does not meet the multi-channel transmission requirements of the capture terminal and rendering terminal in the 3D communication system.

[0004] The existing methods for multi-channel acquisition of three-dimensional motion data can meet the basic requirements, but there are also some potential defects and challenges, which are specifically reflected in the following aspects: First, the existing technology does not pay enough attention to the analysis of the motion capture index and light source compliance index of each character in the film and television production process, which in turn affects the acquisition of motion capture data and optical occlusion data, affects the subjective judgment during the production of traditional manual key frames, and thus leads to motion distortion, which in turn affects the evaluation of the motion smoothness index of each character in the film and television production process, reduces the vitality and appeal of the characters, reduces the artistic expression of film and television works, increases creative differences, reduces decision-making efficiency, increases production cycle, and increases the cost caused by rework.

[0005] Second, the existing technology is not accurate enough in determining whether the action is smooth and matching the lens application level, which reduces the accuracy of lens language and action performance, thereby reducing industrial production efficiency and increasing subjective trial and error time, which in turn affects the realization of refined resource allocation. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for collecting three-dimensional motion data through multiple channels, which solves the problems existing in the background technology.

[0007] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a method for multi-channel acquisition of three-dimensional motion data, including step one, three-dimensional motion data acquisition, step two, inertial measurement, step three, motion analysis and step four, three-dimensional motion judgment.

[0008] Step 1: 3D motion data collection: Use high-definition cameras to collect data on the character movements during film and television production, obtain motion capture data, and then analyze the motion capture index of each character in the film and television production process.

[0009] Step 2: Inertial measurement: Through the matching optical capture system, collect optical occlusion data during the film and television production process, and then analyze the light source compliance index of each character in the film and television production process.

[0010] Step 3: Motion analysis: Based on the obtained motion capture index and light source coincidence index of each character in the film and television production process, the motion smoothness index of each character in the film and television production process is evaluated through the motion smoothness model.

[0011] Step 4: 3D motion judgment: Based on the obtained motion smoothness index of each character in the film and television production process, judge whether the motion is smooth and match the lens application level.

[0012] Furthermore, the data of the character movements in the film and television production process is collected by a high-definition camera, and the specific analysis method is as follows: according to the movements of each character, the position of the physical marker points is collected, and each character wears a marker suit, performs calibration actions, establishes a virtual skeleton posture, defines the skeleton joints according to the principles of human anatomy, establishes a mapping relationship between the virtual skeleton posture and the physical marker points, generates an initial skeleton template, and synchronously shoots the movements of each character through a high-definition camera to capture the two-dimensional coordinates of the marker points at different perspectives to obtain the two-dimensional marker points of different frames, and matches the two-dimensional marker points of different frames to the same physical marker point through the spatial distribution characteristics of the marker points to generate motion capture data.

[0013] Furthermore, the spatial distribution characteristics of the marker points include: angle change rate, maximum distance, minimum distance, instantaneous speed, speed change rate, and smoothness of each marker point.

[0014] Furthermore, the motion capture data includes: position offset values ​​of each marker point of each character in the film and television production process, joint rotation quaternion, linear velocity, angular velocity, and distance between each limb.

[0015] Furthermore, the motion capture index of each character in the film and television production process is specifically analyzed by: based on the obtained motion capture data, the position offset value compliance interval, the joint rotation quaternion compliance interval, the linear velocity compliance interval, the angular velocity compliance interval and the distance compliance interval between each limb of each marker point of each character in the film and television production process are extracted from the database, and the motion capture index of each character in the film and television production process is analyzed. The specific calculation formula is: E ch It represents the hth motion capture data of the dth marker of the cth character in the film and television production process, E ch ' represents the matching interval of the hth motion capture data of the dth marker of the cth character in the film and television production process, h∈[1,5], c represents the number of each character, c=1,2,...,i, i represents the number of characters, d represents the number of each marker, d=1,2,...,f, f represents the number of markers.

[0016] Furthermore, the optical occlusion data specifically includes: the area, angle, distance, occlusion duration, and attenuation gradient of each occlusion area of ​​each character in the film and television production process.

[0017] Furthermore, the light source compliance index of each character in the film and television production process is specifically analyzed by: based on the obtained optical occlusion data, the area compliance interval, the angle compliance interval, the distance compliance interval, the occlusion duration compliance interval, and the attenuation gradient compliance interval of each occlusion area of ​​each character in the film and television production process are extracted from the database, and the light source compliance index of each character in the film and television production process is analyzed. The specific calculation formula is: Q czk It represents the kth optical occlusion data of the zth occlusion area of ​​the cth character in the film and television production process, Q czk ' represents the matching interval of the kth optical occlusion data of the zth occlusion area of ​​the cth character in the film and television production process, k∈[1,5], where z represents the number of the occlusion area, z=1,2,...,a, and a represents the number of occlusion areas.

[0018] Furthermore, the specific analysis method for evaluating the action smoothness index of each character in the film and television production process is as follows: the motion capture index of each character in the film and television production process and the light source coincidence index of each character in the film and television production process are introduced into the action smoothness model, Get the action smoothness index M of each character in the film and television production process c , where ∨ represents the logical symbol or, Represents the logical symbol NOT.

[0019] Furthermore, the specific analysis method for judging whether the action is smooth is as follows: based on the obtained action smoothness index of each character in the film and television production process, the action smoothness index of each character in the film and television production process is compared with the action smoothness index threshold of each character in the film and television production process stored in the database; if the action smoothness index of a character in the film and television production process is less than the action smoothness index threshold of a character in the film and television production process, the action of the character is recorded as a first-level action index, indicating that the action is not smooth and needs to be re-tested; if the action smoothness index of a character in the film and television production process is greater than or equal to the action smoothness index threshold of a character in the film and television production process, the action of the character is recorded as a second-level action smoothness index.

[0020] Furthermore, the matching shot application level extracts the historical shot action reference matching value intervals corresponding to each level of action smoothness index from the database. When the action of a certain character in the film and television production process is a secondary action smoothness index, the secondary action smoothness index is compared with the historical shot action reference matching value interval. If the secondary action smoothness index is in the historical shot action reference matching value interval, the shot matching application level of the character is recorded as a dynamic superior shot, and matched with a high-cost shot. Otherwise, the shot matching application level of the character is recorded as a conventional shot, and matched with a low-cost shot.

[0021] The beneficial effects of the present invention are as follows: in step one, three-dimensional motion data acquisition, step two, inertial measurement and step three, motion analysis: data on the character movements in the film and television production process are collected by a high-definition camera to obtain motion capture data, thereby avoiding motion distortion caused by subjective judgment during traditional manual key frame production, and then analyzing the motion capture index of each character in the film and television production process; through a matching optical capture system, optical occlusion data in the film and television production process is collected, and then the light source compliance index of each character in the film and television production process is analyzed; based on the obtained motion capture index of each character in the film and television production process and the light source compliance index of each character in the film and television production process, the motion smoothness index of each character in the film and television production process is evaluated through a motion smoothness model, thereby improving the vitality and appeal of the characters, enhancing the artistic expression of film and television works, reducing creative differences, improving decision-making efficiency, shortening the production cycle, and reducing costs caused by rework.

[0022] In step 4, three-dimensional motion judgment: based on the obtained motion smoothness index of each character in the film and television production process, judge whether the motion is smooth, match the lens application level, improve the accuracy of lens language and motion performance, improve industrial production efficiency, reduce subjective trial and error time, and achieve refined allocation of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 The figure is a flow chart of the steps for implementing the method of the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] Reference Figure 1 As shown, the present invention provides a method for multi-channel acquisition of three-dimensional motion data, including: step one, three-dimensional motion data acquisition, step two, inertial measurement, step three, motion analysis and step four, three-dimensional motion judgment.

[0027] Step 1: 3D motion data collection: Use high-definition cameras to collect data on the character movements during film and television production, obtain motion capture data, and then analyze the motion capture index of each character in the film and television production process.

[0028] In the above embodiment, the data of the character movements in the film and television production process is collected by a high-definition camera, and the specific analysis method is as follows: according to the movements of each character, the position of the physical marker points is collected, and each character wears a marker suit, performs calibration actions, establishes a virtual skeleton posture, defines the skeleton joints according to the principles of human anatomy, establishes a mapping relationship between the virtual skeleton posture and the physical marker points, generates an initial skeleton template, and synchronously shoots the movements of each character through a high-definition camera, captures the two-dimensional coordinates of the marker points at different perspectives, obtains the two-dimensional marker points of different frames, and matches the two-dimensional marker points of different frames to the same physical marker point through the spatial distribution characteristics of the marker points to generate motion capture data.

[0029] In the above embodiment, the spatial distribution characteristics of the marker points include: the angle change rate, maximum distance, minimum distance, instantaneous speed, speed change rate, and smoothness of each marker point.

[0030] In the above embodiment, the motion capture data includes: position offset values ​​of each marker point of each character in the film and television production process, joint rotation quaternion, linear velocity, angular velocity, and distance between limbs.

[0031] It should be noted that the joint rotation quaternion refers to the pitch angle, yaw angle, roll angle or rotation matrix that represents the orientation of the joint. For example, the rotation quaternion (w, x, y, z) of the elbow joint can describe its bending angle in three-dimensional space.

[0032] It should be noted that linear velocity refers to the displacement change of a marker point per unit time, such as the translation speed of a character's feet when running; angular velocity refers to the speed of joint rotation, such as the rotation rate of the wrist when swinging an arm.

[0033] It should be noted that the distance between the limbs refers to the distance between the hands, which is used to restore the coordination of the movement posture.

[0034] In the above embodiment, the motion capture index of each character in the film and television production process is specifically analyzed by: based on the obtained motion capture data, the position offset value compliance interval of each marker point of each character in the film and television production process, the joint rotation quaternion compliance interval, the linear velocity compliance interval, the angular velocity compliance interval and the distance compliance interval between each limb are extracted from the database, and the motion capture index of each character in the film and television production process is analyzed. The specific calculation formula is: E ch It represents the hth motion capture data of the dth marker of the cth character in the film and television production process, E ch ' represents the matching interval of the hth motion capture data of the dth marker of the cth character in the film and television production process, h∈[1,5], c represents the number of each character, c=1,2,...,i, i represents the number of characters, d represents the number of each marker, d=1,2,...,f, f represents the number of markers.

[0035] Step 2: Inertial measurement: Through the matching optical capture system, collect optical occlusion data during the film and television production process, and then analyze the light source compliance index of each character in the film and television production process.

[0036] In the above embodiment, the optical occlusion data specifically includes: the area, angle, distance, occlusion duration, and attenuation gradient of each occlusion area of ​​each character in the film and television production process.

[0037] It should be noted that the area, angle, and distance of the occlusion of each character are calculated and analyzed through polygonal meshes or point clouds to describe the three-dimensional shape of the occluder, such as the cubic bounding box of the character's torso and the cylindrical model of the arm; the distance of the occlusion of each character refers to the distance from the center of the occluder to the light source; the angle of the occlusion of each character refers to the angle between the surface normal of the occluder and the direction of the light source; the duration of the occlusion refers to the number of consecutive frames that are occluded, such as the character raising his hand to block the light source on his face for 50 frames; the attenuation gradient refers to the lighting transition range at the edge of the occluder, which is obtained through photometer detection, such as the gradient of the soft shadow edge, which is determined by the size of the light source and the distance from the occluder.

[0038] In the above embodiment, the light source compliance index of each character in the film and television production process is specifically analyzed by: based on the obtained optical occlusion data, the area compliance interval, angle compliance interval, distance compliance interval, occlusion duration compliance interval, and attenuation gradient compliance interval of each occlusion area of ​​each character in the film and television production process are extracted from the database, and the light source compliance index of each character in the film and television production process is analyzed. The specific calculation formula is: Q czk It represents the kth optical occlusion data of the zth occlusion area of ​​the cth character in the film and television production process, Q czk ' represents the matching interval of the kth optical occlusion data of the zth occlusion area of ​​the cth character in the film and television production process, k∈[1,5], where z represents the number of the occlusion area, z=1,2,...,a, and a represents the number of occlusion areas.

[0039] Step 3: Motion analysis: Based on the obtained motion capture index and light source coincidence index of each character in the film and television production process, the motion smoothness index of each character in the film and television production process is evaluated through the motion smoothness model.

[0040] In the above embodiment, the specific analysis method for evaluating the motion smoothness index of each character in the film and television production process is as follows: the motion capture index of each character in the film and television production process and the light source coincidence index of each character in the film and television production process are introduced into the motion smoothness model, Get the action smoothness index M of each character in the film and television production process c , where ∨ represents the logical symbol or, Represents the logical symbol NOT.

[0041] In step one, three-dimensional motion data acquisition, step two, inertial measurement, and step three, motion analysis: data on the character movements in the film and television production process is collected through high-definition cameras to obtain motion capture data, thereby avoiding motion distortion caused by subjective judgment during traditional manual key frame production, and then analyzing the motion capture index of each character in the film and television production process. Through the matching optical capture system, optical occlusion data in the film and television production process is collected, and then the light source compliance index of each character in the film and television production process is analyzed. Based on the obtained motion capture index and light source compliance index of each character in the film and television production process, the motion smoothness index of each character in the film and television production process is evaluated through a motion smoothness model, thereby improving the vitality and appeal of the characters, enhancing the artistic expression of film and television works, reducing creative differences, improving decision-making efficiency, shortening the production cycle, and reducing costs caused by rework.

[0042] Step 4: 3D motion judgment: Based on the obtained motion smoothness index of each character in the film and television production process, judge whether the motion is smooth and match the lens application level.

[0043] In the above embodiment, the specific analysis method for judging whether the action is smooth is: based on the obtained action smoothness index of each character in the film and television production process, the action smoothness index of each character in the film and television production process is compared with the action smoothness index threshold of each character in the film and television production process stored in the database. If the action smoothness index of a character in the film and television production process is less than the action smoothness index threshold of a character in the film and television production process, the action of the character is recorded as a first-level action index, indicating that the action is not smooth and needs to be re-tested. If the action smoothness index of a character in the film and television production process is greater than or equal to the action smoothness index threshold of a character in the film and television production process, the action of the character is recorded as a second-level action smoothness index.

[0044] In the above embodiment, the matching shot application level extracts the historical shot action reference matching value interval corresponding to each level of action smoothness index from the database. When the action of a certain character in the film and television production process is a secondary action smoothness index, the secondary action smoothness index is compared with the historical shot action reference matching value interval. If the secondary action smoothness index is in the historical shot action reference matching value interval, the shot matching application level of the character is recorded as a dynamic superior shot, and the high-cost shot is matched. Otherwise, the shot matching application level of the character is recorded as a conventional shot, and the low-cost shot is matched.

[0045] In step 4, three-dimensional motion judgment: based on the obtained motion smoothness index of each character in the film and television production process, judge whether the motion is smooth, match the lens application level, improve the accuracy of lens language and motion performance, improve industrial production efficiency, reduce subjective trial and error time, and achieve refined allocation of resources.

[0046] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. A method for collecting three-dimensional motion data through multiple channels, characterized in that: include: Step 1: 3D motion data collection: Use high-definition cameras to collect data on the character movements during film and television production, obtain motion capture data, and then analyze the motion capture index of each character in the film and television production process; Step 2: Inertial measurement: Using an optical capture system, we collect optical occlusion data during film and television production, and then analyze the light source compliance index of each character in the film and television production process; Step 3: Motion Analysis: Based on the obtained motion capture index and light source coincidence index of each character in the film and television production process, the motion smoothness index of each character in the film and television production process is evaluated through the motion smoothness model; Step 4: 3D motion judgment: Based on the obtained motion smoothness index of each character in the film and television production process, judge whether the motion is smooth and match the lens application level.

2. The method for collecting three-dimensional motion data through multiple channels according to claim 1, wherein: The specific analysis method for collecting data on character movements during film and television production using a high-definition camera is as follows: According to the actions of each character, the position of the physical marker points is collected, and each character is equipped with a marker suit to perform calibration actions, establish a virtual skeleton posture, define the bone joints according to the principles of human anatomy, establish a mapping relationship between the virtual skeleton posture and the physical marker points, generate an initial skeleton template, and synchronously shoot the actions of each character through a high-definition camera to capture the two-dimensional coordinates of the marker points at different perspectives, obtain the two-dimensional marker points of different frames, and match the two-dimensional marker points of different frames to the same physical marker point through the spatial distribution characteristics of the marker points to generate motion capture data.

3. The method for collecting three-dimensional motion data through multiple channels according to claim 2, wherein: The spatial distribution characteristics of the marker points include: the angle change rate, maximum distance, minimum distance, instantaneous speed, speed change rate, and smoothness of each marker point.

4. The method for collecting three-dimensional motion data through multiple channels according to claim 3, wherein: The motion capture data includes: position offset values ​​of each marker point of each character in the film and television production process, joint rotation quaternion, linear velocity, angular velocity, and distance between limbs.

5. The method for collecting three-dimensional motion data through multiple channels according to claim 4, characterized in that: The specific analysis method for the motion capture index of each character in the film and television production process is as follows: Based on the obtained motion capture data, the position offset value of each marker point of each character in the film and television production process is extracted from the database to meet the interval, the joint rotation quaternion meets the interval, the linear velocity meets the interval, the angular velocity meets the interval, and the distance between each limb meets the interval. The motion capture index of each character in the film and television production process is analyzed. The specific calculation formula is: E ch It represents the hth motion capture data of the dth marker of the cth character in the film and television production process, E ch ' represents the matching interval of the hth motion capture data of the dth marker of the cth character in the film and television production process, h∈[1,5], c represents the number of each character, c=1,2,...,i, i represents the number of characters, d represents the number of each marker, d=1,2,...,f, f represents the number of markers.

6. The method for collecting three-dimensional motion data through multiple channels according to claim 1, wherein: The optical occlusion data specifically includes: the area, angle, distance, occlusion duration, and attenuation gradient of each occlusion area of ​​each character in the film and television production process.

7. The method for collecting three-dimensional motion data through multiple channels according to claim 6, characterized in that: The light source compliance index of each character in the film and television production process is analyzed in the following specific way: Based on the obtained optical occlusion data, the area compliance interval, angle compliance interval, distance compliance interval, occlusion duration compliance interval, and attenuation gradient compliance interval of each occlusion area of ​​each character in the film and television production process are extracted from the database to analyze the light source compliance index of each character in the film and television production process. The specific calculation formula is: Q czk It represents the kth optical occlusion data of the zth occlusion area of ​​the cth character in the film and television production process, Q czk ' represents the matching interval of the kth optical occlusion data of the zth occlusion area of ​​the cth character in the film and television production process, k∈[1,5], where z represents the number of the occlusion area, z=1,2,...,a, and a represents the number of occlusion areas.

8. The method for collecting three-dimensional motion data through multiple channels according to claim 1, wherein: The specific analysis method for evaluating the action fluency index of each character in the film and television production process is as follows: Import the motion capture index of each character in the film and television production process and the light source index of each character in the film and television production process into the motion smoothness model, Get the action smoothness index M of each character in the film and television production process c , where ∨ represents the logical symbol or, Represents the logical symbol NOT.

9. The method for collecting three-dimensional motion data through multiple channels according to claim 1, wherein: The specific analysis method for determining whether the action is smooth is as follows: Based on the obtained action fluency index of each character in the film and television production process, the action fluency index of each character in the film and television production process is compared with the action fluency index threshold of each character in the film and television production process stored in the database. If the action fluency index of a character in the film and television production process is less than the action fluency index threshold of a character in the film and television production process, the action of the character is recorded as a first-level action index, indicating that the action is not smooth and needs to be re-tested. If the action fluency index of a character in the film and television production process is greater than or equal to the action fluency index threshold of a character in the film and television production process, the action of the character is recorded as a second-level action fluency index.

10. The method for collecting three-dimensional motion data through multiple channels according to claim 1, wherein: The matching lens application level, The historical lens action reference matching value intervals corresponding to each level of action smoothness index are extracted from the database. When the action of a certain character in the film and television production process is a secondary action smoothness index, the secondary action smoothness index is compared with the historical lens action reference matching value interval. If the secondary action smoothness index is within the historical lens action reference matching value interval, the lens matching application level of the character is recorded as a dynamic superior lens, and is matched with a high-cost lens. Otherwise, the lens matching application level of the character is recorded as a conventional lens, and is matched with a low-cost lens.

Images