Application method for driving artificial intelligence module to realize automatic fusion of advertisement pictures

Abstract

The invention discloses an application method for driving an artificial intelligence module to realize automatic fusion of advertisement pictures, and the method employs an independent artificial intelligence module to work, and the artificial intelligence module is provided with computing power and high-speed storage, is connected with a studio host through a high-speed interface, receives the unified scheduling and task distribution of the studio host, and achieves the automatic fusion of the advertisement pictures. The artificial intelligence module belongs to an edge calculation and reasoning unit, a pre-training model or a weighing file is used for working, and the application method for automatically fusing the advertisement pictures is divided into three stages, namely, an A stage, a B stage and a C stage. Advanced dynamic studio and random composite studio functions can be provided, the efficiency and diversified requirements of the studio system are greatly improved, the display processing requirement of the host studio system is effectively reduced by using the artificial intelligence module, the broadcast control cost is saved, the upper manual processing link is reduced, and the efficiency of the host studio system is improved. According to the method, the playing safety is effectively enhanced, beneficial edge calculation and decentralization processes are realized, and compared with an existing advertisement playing mode, the improvement effect is obvious.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an application method for driving an artificial intelligence module to realize automatic fusion of advertising images. Background Art

[0002] At present, with the advancement of display technology, the technological boundaries of the advertising industry have also expanded, and various broadcasting and control systems have been continuously updated. However, in video advertising, especially outdoor video advertising media, various screens of unconventional sizes and shapes require a lot of resources for program source production, adaptation, and management. In addition to the high operation and maintenance costs, the results are often unable to meet all the needs and unique creativity of advertising users. For this reason, we propose an application method that drives an artificial intelligence module to realize automatic fusion of advertising images. Summary of the invention

[0003] The purpose of the present invention is to provide an application method for driving an artificial intelligence module to realize automatic fusion of advertising images, so as to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an application method for driving an artificial intelligence module to realize automatic fusion of advertising images, the method adopts an independent artificial intelligence module, the artificial intelligence module has its own computing power and high-speed storage, and is connected to the broadcast host through a high-speed interface, and accepts the unified scheduling and task allocation of the broadcast host.

[0005] Preferably, the artificial intelligence module belongs to the edge computing and reasoning unit, and works using a pre-trained model or weight file.

[0006] Preferably, the application method of automatic fusion of advertising images is divided into three stages, ABC: A simulation sketch stage, B video factory stage and C dynamic script stage, and the corresponding files or scripts are inferred and generated respectively.

[0007] Preferably, in the A simulation sketch stage, the work of the artificial intelligence module is to generate a simulated screen sketch file or screen layout file and a vectorized fusion script, wherein the simulated screen sketch file is a previewable working file similar to Sketch data, which only borrows the expression mode of the Sketch data source and is not Sketch data. It can be quickly previewed on a virtual screen or a real screen, and is also used for system or background verification and inspection of working conditions. The vectorized fusion script is used to generate a simple rendered verification bitmap and screen material processing package that expresses the placement and order of the material.

[0008] Preferably, in the B video factory stage, the work of the artificial intelligence module is to dynamically generate a fused video file of the actual advertising screen required by the broadcasting system. In the B video factory stage, one or more simulation screen sketches can be applied to fuse and generate a layout effect video of the same advertisement. The A simulation sketch stage can also be dynamically scheduled according to demand to quickly generate a new layout sketch, and then proceed to the B video factory stage.

[0009] Preferably, in the C dynamic script stage, the work of the artificial intelligence module is to generate another real-time picture fusion broadcasting method, that is, a dynamic vectorized broadcasting method. At the same time, the C dynamic script stage can also call the video file generated by the B video factory stage to reorganize the layout or call back the A simulation sketch stage for iterative construction.

[0010] Preferably, the studio system or studio host can parallelize the scheduling of artificial intelligence modules at any time. After the first complete scheduling of A simulation sketch stage, B video factory stage and C dynamic script stage, it can perform quick scheduling of any single stage at any time, and can also perform complete scheduling again at any time.

[0011] Compared with the prior art, the present invention has the following beneficial effects: The present invention can provide advanced dynamic broadcasting and random composite broadcasting functions, greatly improving the efficiency and diversified needs of the broadcasting system, and using the artificial intelligence module to effectively reduce the display processing requirements of the host broadcasting system, saving broadcast control costs, reducing upper-level manual processing links, effectively enhancing broadcasting safety, and realizing beneficial edge computing and decentralized processes. Compared with the existing advertising playback mode, the improvement effect is obvious. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Attached Figure 1 This is a flow chart of the automatic fusion stage A (simulation sketch) of the present invention for implementing the advertising screen; Attached Figure 2 This is a flow chart of the automatic fusion stage B (video factory) of the present invention for implementing the advertisement screen; Attached Figure 3 This is a flow chart of the present invention for realizing the automatic fusion of advertising images in stage C (dynamic script); Attached Figure 4 It is a comparison chart between conventional advertising screen production, delivery, and broadcasting processes and methods and the application method and broadcasting method of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0014] See also Figure 1-Figure 4 An application method for driving an artificial intelligence module to realize automatic fusion of advertising images. The method adopts an independent artificial intelligence module. The artificial intelligence module has its own computing power and high-speed storage, and is connected to a studio host through a high-speed interface (such as Mini PCIe, M.2 or SO-DIMM), and accepts unified scheduling and task allocation of the studio host.

[0015] The artificial intelligence module belongs to the edge computing and reasoning unit, and uses a pre-trained model or weight file to work. The pre-trained model or weight file is the special deep learning and training result of the upper advanced computing platform or cloud computing system. This patent is a special application method for edge computing and terminals.

[0016] The application method of automatic fusion of advertising images is divided into three stages: A, B, and C: A simulation sketch stage, B video factory stage, and C dynamic script stage, and the corresponding files or scripts are inferred and generated respectively.

[0017] In the A simulation sketch stage, the work of the artificial intelligence module is to generate a simulated screen sketch file or screen layout file and a vector fusion script. The simulated screen sketch file is a previewable working file similar to Sketch data. It only borrows the expression mode of the Sketch data source and is not Sketch data. It can be quickly previewed on a virtual screen or a real screen. It is also used for system or background (including manual maintenance background) verification and inspection of working conditions. The vector fusion script is used to generate a simple rendering verification bitmap (BITMAP) and screen material processing package that expresses the placement and order of the material.

[0018] In the B video factory stage, the job of the artificial intelligence module is to dynamically generate a fusion video file of the actual advertising screen required by the studio system. In the B video factory stage, one or more simulation screen sketches can be applied to fuse and generate the layout effect video of the same advertisement. The A simulation sketch stage can also be dynamically scheduled according to demand to quickly generate a new layout sketch, and then proceed to the B video factory stage. The B video factory stage is very suitable for complex layouts, multi-element screens, advanced video effects presentation, and the need to crop, split, and reorganize ultra-large, ultra-long, and ultra-wide screens, as well as scenarios that require hardware encoding and decoding.

[0019] In the C dynamic script stage, the work of the artificial intelligence module is to generate another real-time picture fusion broadcasting method, that is, the dynamic vectorized broadcasting method. Compared with the video generation method in the B video factory stage, the vectorized broadcasting method has the advantages of less processing resource usage, fast scheduling speed, the ability to correct or adjust the picture at any time during playback, and the ability to add other linked APP response functions. It is very suitable for scenes with simple main material pictures, few layout changes, and no picture gradient requirements, as well as quickly meeting the needs of multi-screen picture interaction, linkage, delay and synchronous broadcasting. At the same time, the C dynamic script stage can also call the video files generated in the B video factory stage to reorganize the layout or call back the A simulation sketch stage for iterative construction.

[0020] The studio system or studio host can parallelize and schedule (without waiting) the artificial intelligence module at any time. After the first complete scheduling of the A simulation sketch stage, B video factory stage and C dynamic script stage, it can perform quick scheduling of any single stage at any time (instant fine-tuning) and re-schedule the entire stage at any time (iterative construction), thus truly realizing dynamic studio.

[0021] Different from the conventional advertising screen delivery and broadcasting process, this patent does not need to invest a lot of manpower and material resources on the upper end for advertising customization and program production. Instead, the main advertising materials (video graphics, etc.) and other materials (optional: patterns, introductions, color tables, fonts, logos, constraint words, etc.) are directly delivered to the broadcasting terminal, and the broadcasting terminal drives the artificial intelligence module (artificial intelligence module) to automatically process the advertising materials. This method is suitable for terminal or edge processing of various advertising broadcasting screens, and can automatically merge and generate adaptive advertising programs to meet various broadcasting needs, including but not limited to: automatic fitting of various unconventional screens, special-shaped screens, extra-long or extra-wide screens, dynamic fusion of graphics and subtitles, automatic content completion, dynamic insertion of time components, automatic identification and separation of key elements, automatic multi-screen splicing, automatic arrangement or rearrangement of videos and graphics, generative content filling of blank screens, automatic removal or automatic modification of black edges, etc.

[0022] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An application method for driving an artificial intelligence module to realize automatic fusion of advertising images, characterized in that: The method adopts an independent artificial intelligence module, which has its own computing power and high-speed storage, is connected to the studio host through a high-speed interface, and accepts unified scheduling and task allocation from the studio host.

2. The application method of driving an artificial intelligence module to realize automatic fusion of advertising images according to claim 1, characterized in that: The artificial intelligence module belongs to the edge computing and reasoning unit and works using pre-trained models or weight files.

3. The application method of driving an artificial intelligence module to realize automatic fusion of advertising images according to claim 1, characterized in that: The application method of automatic fusion of advertising images is divided into three stages: A, B, and C: A simulation sketch stage, B video factory stage, and C dynamic script stage, and the corresponding files or scripts are inferred and generated respectively.

4. The application method of driving an artificial intelligence module to realize automatic fusion of advertising images according to claim 3, characterized in that: In the A simulation sketch stage, the work of the artificial intelligence module is to generate a simulated screen sketch file or screen layout file and a vector fusion script. The simulated screen sketch file is a previewable working file similar to Sketch data. It only borrows the expression mode of the Sketch data source and is not Sketch data. It can be quickly previewed on a virtual screen or a real screen. It is also used for system or background verification and inspection of working conditions. The vector fusion script is used to generate a simple rendering verification bitmap and screen material processing package that expresses the placement and order of the material.

5. The application method of driving an artificial intelligence module to realize automatic fusion of advertising images according to claim 3, characterized in that: In the B video factory stage, the job of the artificial intelligence module is to dynamically generate a fusion video file of the actual advertising images required by the broadcasting system. In the B video factory stage, one or more simulation picture sketches can be applied to fuse and generate the layout effect video of the same advertisement. The A simulation sketch stage can also be dynamically scheduled according to demand to quickly generate a new layout sketch, and then proceed to the B video factory stage.

6. The application method of driving an artificial intelligence module to realize automatic fusion of advertising images according to claim 3, characterized in that: In the C dynamic script stage, the work of the artificial intelligence module is to generate another real-time picture fusion broadcasting method, that is, the dynamic vectorized broadcasting method. At the same time, the C dynamic script stage can also call the video file generated by the B video factory stage to reorganize the layout or call back the A simulation sketch stage for iterative construction.

7. The application method of driving an artificial intelligence module to realize automatic fusion of advertising images according to claim 5, characterized in that: The studio system or studio host can parallelize the scheduling of artificial intelligence modules at any time. After the first complete scheduling of A simulation sketch stage, B video factory stage and C dynamic script stage, it can perform quick scheduling of any single stage at any time, and can also perform complete scheduling again at any time.

Images