Search

CN-121999186-A - Augmented reality art display system

CN121999186ACN 121999186 ACN121999186 ACN 121999186ACN-121999186-A

Abstract

The invention relates to the technical field of augmented reality and computer vision processing, in particular to an augmented reality artistic display system which comprises an environment perception processing center, an intention constraint unit and an intention constraint unit, wherein the environment perception processing center is used for calling multidimensional environment sensing data to obtain environment illumination characteristics and scene semantic characteristics, coupling processing is carried out on the environment illumination characteristics and the scene semantic characteristics to obtain luminosity semantic coupling data, when luminosity semantic coupling data are generated, an imbalance prediction unit is used for obtaining an aesthetic imbalance index, the intention constraint unit is used for obtaining a conventional rendering signal or a reconstruction rendering signal, when the reconstruction rendering signal is generated, the reconstruction rendering unit is used for carrying out vision consistency synthesis on an obtained compensation model and a real video stream to obtain an augmented reality display image.

Inventors

  • DAI XINYU

Assignees

  • 浙江财经大学

Dates

Publication Date
20260508
Application Date
20260119

Claims (8)

  1. 1. The augmented reality art display system is characterized by comprising an environment perception processing center, a luminosity semantic analysis unit, an imbalance prediction unit, an intention constraint unit and a reconstruction rendering unit; The environment sensing processing center is used for calling multi-dimensional environment sensing data, sending the multi-dimensional environment sensing data to the luminosity semantic analysis unit for extracting and analyzing environment characteristics to obtain environment illumination characteristics and scene semantic characteristics, and performing coupling processing on the environment illumination characteristics and the scene semantic characteristics to obtain luminosity semantic coupling data; When the luminosity semantic coupling data are generated, the imbalance prediction unit is used for carrying out aesthetic matching measurement and calculation analysis on the collected artistic intention descriptors of the virtual artwork to obtain an aesthetic imbalance index; the intention constraint unit is used for carrying out threshold judgment processing on the trial imbalance index to obtain a conventional rendering signal or a reconstructed rendering signal; When a reconstruction rendering signal is generated, the reconstruction rendering unit is used for carrying out self-adaptive compensation reconstruction analysis on rendering parameters of the virtual artwork, and carrying out visual consistency synthesis on the obtained compensation model and the real video stream to obtain an augmented reality display image.
  2. 2. The augmented reality artistic display system according to claim 1, wherein the environmental feature extraction and analysis process is as follows, a real environment is collected Video stream data and will Setting video stream data as a visual input source, performing spherical harmonic coefficient calculation on illumination information in the visual input source to obtain main light source direction data and environment light color temperature data, setting the main light source direction data and the environment light color temperature data as environment illumination characteristics, performing semantic segmentation processing on the visual input source, identifying plane attribute data and object attribute data, and setting the plane attribute data and the object attribute data as scene semantic characteristics.
  3. 3. The augmented reality art display system according to claim 2 is characterized in that the aesthetic matching, measuring and calculating and analyzing process comprises the steps of obtaining an art intention descriptor of a virtual artwork, comparing and analyzing an environment illumination characteristic with a preset ideal illumination parameter to obtain an illumination deviation value, comparing and analyzing a scene semantic characteristic with the preset ideal layout parameter to obtain a semantic conflict value, and setting a product value obtained by multiplying the illumination deviation value and the semantic conflict value by a numerical value after data normalization processing to be an aesthetic disorder index.
  4. 4. The augmented reality art display system of claim 1, wherein the threshold discrimination process is configured to compare an aesthetic imbalance index to a preset imbalance threshold, generate a conventional rendering signal for instructing the system to maintain original rendering parameters of the virtual artwork if the aesthetic imbalance index is less than the preset imbalance threshold, and generate a reconstructed rendering signal for triggering an adaptive compensation reconstruction analysis if the aesthetic imbalance index is greater than or equal to the preset imbalance threshold.
  5. 5. The augmented reality art display system according to claim 4, wherein the adaptive compensation reconstruction analysis process is characterized in that when a reconstruction rendering signal is generated, a dominant factor causing an over-standard aesthetic disorder index is obtained, the dominant factor is an illumination deviation or a semantic conflict, if the dominant factor is the illumination deviation, tone mapping inverse compensation processing is performed on texture of a virtual artwork to generate model data after illumination compensation, and if the dominant factor is the semantic conflict, morphological fine adjustment or background mask generation processing is performed on geometric forms of the virtual artwork to generate model data after morphological reconstruction.
  6. 6. The augmented reality art display system of claim 5, wherein the tone mapping reverse compensation process comprises obtaining a current color temperature value of the real environment, calculating a difference between the current color temperature value and an intrinsic color temperature value of the virtual artwork to obtain a color temperature drift amount, generating a reverse color filter matrix based on the color temperature drift amount, and applying the reverse color filter matrix to a texture map of the virtual artwork to maintain a preset perceived color of the virtual artwork under real environment illumination.
  7. 7. An augmented reality art display system according to claim 3, wherein the illumination deviation value represents a sine value of an angle difference between a main light source direction of the real environment and a virtual artwork preset shadow casting direction, and the semantic conflict value represents a product value obtained by multiplying a portion of a pixel ratio of a clutter object in the real environment in a field of view exceeding a preset background purity threshold value by a value obtained by carrying out data normalization processing on a blocked area ratio of the virtual artwork.
  8. 8. The augmented reality art display system of claim 1, wherein the adaptive compensation reconstruction analysis further comprises a differential rendering process of generating a transparent layer containing only shadow information and reflection information, setting the transparent layer as an environmental fusion layer, overlaying the virtual artwork model after the compensation reconstruction onto the environmental fusion layer, and adjusting soft and hard parameters of shadow casting according to planar attribute data in semantic features of the scene to generate a final compensation model.

Description

Augmented reality art display system Technical Field The invention relates to the technical field of augmented reality and computer vision processing, in particular to an augmented reality artistic display system. Background In the current augmented reality art display technology field, the system needs to superimpose virtual artwork into the real video stream to present immersive visual effects. The method is characterized in that a geometric position-based registration technology is generally adopted, the positioning accuracy and tracking stability of a virtual model in a three-dimensional space coordinate system are focused, the position accuracy of the model can be guaranteed, however, the method ignores the interference of complex illumination changes, background semantic confusion and other factors on aesthetic intention of artworks, so that under an uncontrolled environment, visual cracking sense and terrorist valley effect are often generated by a virtual object due to inconsistent light and shadow or background conflict, and the integrity of artistic expression is seriously damaged, therefore, a quantification mechanism of the aesthetic perception of the environment is established, and self-adaptive visual compensation of the virtual artworks is realized, so that aesthetic consistency and immersion sense under an augmented reality scene are improved, and the technical problem to be solved is urgently. Disclosure of Invention In order to solve the technical problems, the invention provides an augmented reality art display system, which specifically comprises the following technical scheme: the system comprises an environment perception processing center, a luminosity semantic analysis unit, an imbalance prediction unit, an intention constraint unit and a reconstruction rendering unit; The environment sensing processing center is used for calling multi-dimensional environment sensing data, sending the multi-dimensional environment sensing data to the luminosity semantic analysis unit for extracting and analyzing environment characteristics to obtain environment illumination characteristics and scene semantic characteristics, and performing coupling processing on the environment illumination characteristics and the scene semantic characteristics to obtain luminosity semantic coupling data; When the luminosity semantic coupling data are generated, the imbalance prediction unit is used for carrying out aesthetic matching measurement and calculation analysis on the collected artistic intention descriptors of the virtual artwork to obtain an aesthetic imbalance index; the intention constraint unit is used for carrying out threshold judgment processing on the trial imbalance index to obtain a conventional rendering signal or a reconstructed rendering signal; When a reconstruction rendering signal is generated, the reconstruction rendering unit is used for carrying out self-adaptive compensation reconstruction analysis on rendering parameters of the virtual artwork, and carrying out visual consistency synthesis on the obtained compensation model and the real video stream to obtain an augmented reality display image. Optionally, the environment characteristic extraction and analysis process comprises collecting real environmentVideo stream data and willSetting video stream data as a visual input source, performing spherical harmonic coefficient calculation on illumination information in the visual input source to obtain main light source direction data and environment light color temperature data, setting the main light source direction data and the environment light color temperature data as environment illumination characteristics, performing semantic segmentation processing on the visual input source, identifying plane attribute data and object attribute data, and setting the plane attribute data and the object attribute data as scene semantic characteristics. The aesthetic matching measuring and calculating analysis process comprises the steps of obtaining an artistic intention descriptor of a virtual artwork, wherein the artistic intention descriptor comprises a preset ideal illumination parameter and a preset ideal layout parameter, comparing and analyzing an environment illumination characteristic with the preset ideal illumination parameter to obtain an illumination deviation value, comparing and analyzing a scene semantic characteristic with the preset ideal layout parameter to obtain a semantic conflict value, and setting a product value obtained by multiplying the illumination deviation value and the semantic conflict value by a numerical value after data normalization processing as an aesthetic disorder index. Optionally, the threshold value judging and processing process includes comparing and analyzing the aesthetic disorder index with a preset disorder threshold value, generating a conventional rendering signal for indicating the system to keep original rendering parameters of the virtual artw