Image generation method and image generation system using the same

Abstract

An image generation method and an image generation system using the same are disclosed. The image generation method includes the following steps: Executing a text-to-image program to generate a base image based on a text prompt; Executing an image expansion drawing program to obtain an expanded image based on the base image; Executing a resolution adjustment program to adjust the resolution of the expanded image to obtain an image with adjusted resolution; Executing a hidden watermark compositing program to composite a hidden watermark onto the image with adjusted resolution to obtain a generated image. The text-to-image program, the image expansion drawing program, the resolution adjustment program, and the hidden watermark compositing program are executed in a pipelined manner through a graphics processor and a neural processor.

Description

Technical Field

[0001] This disclosure relates to an artificial intelligence method and an artificial intelligence system applying the same, and more particularly to an image generation method and an image generation system applying the same. Prior Technology

[0002] With the rapid development of artificial intelligence technology, various image generation and image modification techniques have emerged. However, currently, if multiple programs need to be executed using artificial intelligence technology, a huge amount of system resources must be consumed to achieve this. For example, an NVIDIA 2080 16GB processor is required to generate a 1280x720 AI image, which makes it difficult for artificial intelligence technology to be widely adopted in various laptops.

[0003] Therefore, effectively reducing system resources and shortening image processing time is the direction that the industry is currently striving to address. Summary of the Invention

[0004] This disclosure relates to an image generation method and an image generation system using the same, which utilizes a pipeline approach to execute text-to-image conversion, image expansion and rendering, resolution adjustment, and hidden watermark synthesis processes, thereby significantly improving the speed of image generation.

[0005] According to one aspect of this disclosure, an image generation method is proposed. The image generation method includes the following steps: Executing a text-to-image program to generate a base image based on a text prompt. Executing an image expansion drawing program to obtain an expanded image based on the base image. Executing a resolution adjustment program to adjust the resolution of the expanded image to obtain an image with adjusted resolution. Executing a hidden watermark compositing program to composite a hidden watermark onto the image with adjusted resolution to obtain a generated image. The text-to-image program, image expansion drawing program, resolution adjustment program, and hidden watermark compositing program are executed in a pipelined manner using a Graphics Processing Unit (GPU) and a Neural Network Processing Unit (NPU).

[0006] According to another aspect of this disclosure, an image generation system is proposed. The image generation system includes a text-to-image module, an image expansion and rendering module, a resolution adjustment module, and a hidden watermark synthesis module. The text-to-image module executes a text-to-image program to generate a base image based on a text prompt. The image expansion and rendering module executes an image expansion and rendering program to obtain an expanded image based on the base image. The resolution adjustment module executes a resolution adjustment program to adjust the resolution of the expanded image to obtain an image with adjusted resolution. The hidden watermark synthesis module executes a hidden watermark synthesis program to synthesize a hidden watermark on the image with adjusted resolution to obtain a generated image. The text-to-image program, the image expansion and rendering program, the resolution adjustment program, and the hidden watermark synthesis program are executed in a pipelined manner using a Graphics Processing Unit (GPU) and a Neural Network Processing Unit (NPU).

[0007] According to another aspect of this disclosure, an image generation method is proposed. The image generation method includes the following steps: Executing a text-to-image program to generate a base image based on a text prompt. Executing an image expansion drawing program to obtain an expanded image based on the base image. Executing a resolution adjustment program to adjust the resolution of the expanded image to obtain an image with adjusted resolution. Executing a hidden watermark synthesis program to synthesize a hidden watermark on the image with adjusted resolution to obtain a generated image. In the image expansion drawing program, resolution adjustment program, and hidden watermark synthesis program, intermediate computation results are directly obtained from a memory of a Graphics Processing Unit (GPU) or a memory of a Neural Network Processing Unit (NPU).

[0008] To provide a better understanding of the above and other aspects of this disclosure, specific embodiments are described below in conjunction with the accompanying drawings: Simple Explanation of the Diagram

[0009] Figure 1 illustrates an image generation method according to an embodiment of this disclosure. Figure 2 illustrates a pipelined process for executing text-to-image conversion, image expansion drawing, resolution adjustment, and watermark hiding / compositing. Figure 3 illustrates a block diagram of an image generation system according to an embodiment of the present disclosure. Figure 4 illustrates a flowchart of an image generation method according to an embodiment of the present disclosure. Figure 5 illustrates the image extension drawing procedure. Figure 6 illustrates the relationship between the preprocessed image and the image to be drawn. Figure 7 illustrates one way in which the image generation method disclosed herein achieves zero-copying. Figure 8 illustrates another way in which the image generation method disclosed herein achieves zero-copying. Implementation

[0010] The technical terms used in this specification are based on common terminology in the field. Where this specification provides explanations or definitions for certain terms, the interpretation of those terms shall be based on the explanations or definitions provided in this specification. Each of the embodiments disclosed herein has one or more technical features. Where feasible, those skilled in the art may selectively implement some or all of the technical features in any embodiment, or selectively combine some or all of the technical features in these embodiments.

[0011] Please refer to Figure 1, which illustrates an image generation method according to an embodiment of this disclosure. The image generation method includes, for example, a text-to-image program PD1, an image expansion and rendering program PD2, a resolution adjustment program PD3, and a hidden watermark compositing program PD4. In the text-to-image program PD1, a base image IM1 is generated based on text prompts PT using artificial intelligence (AI) technology. Next, in the image expansion and rendering program PD2, the base image IM1 is expanded and extended outwards to obtain an expanded image IM8. Then, in the resolution adjustment program PD3, the resolution of the expanded image IM8 is adjusted to obtain a resolution-adjusted image IM9. Next, in the hidden watermark compositing program PD4, a hidden watermark WM is added to the expanded image IM8 to obtain a generated image IM10. The hidden watermark WM is scattered across several pixels of the generated image, and its presence is imperceptible to the human eye.

[0012] Please refer to Figure 2, which illustrates a pipelined execution of the text-to-image program PD1, the image expansion and rendering program PD2, the resolution adjustment program PD3, and the hidden watermark synthesis program PD4. In this embodiment, the four programs—text-to-image program PD1, image expansion and rendering program PD2, resolution adjustment program PD3, and hidden watermark synthesis program PD4—are executed in a pipelined manner, for example, through a Graphics Processing Unit (GPU) and a Neural Network Processing Unit (NPU).

[0013] For example, as shown in Figure 2, from time point T1 to time point T2, the graphics processing unit (GPU) can process the text-to-image program PD1 of the text prompt PT1.

[0014] Next, as shown in Figure 2, from time point T2 to time point T3, the neural processor NPU can process the image extension drawing program PD2 of the text prompt PT1, and the graphics processor GPU can process the text-to-image program PD1 of the text prompt PT2.

[0015] Then, as shown in Figure 2, from time point T3 to time point T4, the neural processor NPU can process the resolution adjustment program PD3 of text prompt PT1 and the image expansion drawing program PD2 of text prompt PT2, and the graphics processor GPU can process the text-to-image program PD1 of text prompt PT3.

[0016] Next, as shown in Figure 2, from time point T4 to time point T5, the GPU processes the hidden watermark synthesis procedure PD4 for text prompt PT1, the image extension drawing procedure PD2 for text prompt PT3, and the resolution adjustment procedure PD3 for text prompt PT3. Subsequent time points T6 and T7 follow the same pattern to obtain the generated images IM101, IM102, and IM103 at each time point.

[0017] In other words, in this embodiment, the image generation method is executed in a pipelined manner, which can utilize the graphics processing unit (GPU) and neural processing unit (NPU) to process multiple programs simultaneously, thereby accelerating the processing speed.

[0018] Please refer to Figure 3, which illustrates a block diagram of an image generation system 1000 according to an embodiment of this disclosure. The image generation system 1000 includes a text-to-image module 100, an image expansion and rendering module 200, a resolution adjustment module 300, and a hidden watermark compositing module 400. The text-to-image module 100 is used to convert text into an image. The image expansion and rendering module 200 is used to expand the image. The resolution adjustment module 300 is used to adjust the image resolution. The hidden watermark compositing module 400 is used to add a watermark to the image.

[0019] As shown in Figure 3, the image expansion drawing module 200 includes a preprocessing unit 220, a cutting unit 230, a masking unit 240, an expansion unit 250, a noise reduction unit 260, and a stitching unit 280. The text-to-image module 100, the image expansion drawing module 200, the preprocessing unit 220, the cutting unit 230, the masking unit 240, the expansion unit 250, the noise reduction unit 260, the stitching unit 280, the resolution adjustment module 300, and the hidden watermark synthesis module 400 are, for example, a circuit, a circuit board, a storage device for stored code, or a chip. The chip may be a processor, or other programmable general-purpose or special-purpose microcontroller (MCU), microprocessor, digital signal processor (DSP), programmable controller, application-specific integrated circuit (ASIC), image signal processor (ISP), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field programmable gate array (FPGA), or other similar components or combinations thereof. The text-to-image module 100, image expansion drawing module 200, resolution adjustment module 300, and hidden watermark synthesis module 400 of the image generation system 1000 respectively execute the text-to-image program PD1, image expansion drawing program PD2, resolution adjustment program PD3, and hidden watermark synthesis program PD4. A flowchart is provided below to explain the operation of each component in detail.

[0020] Please refer to Figures 3 and 4 simultaneously. Figure 4 illustrates a flowchart of an image generation method according to an embodiment of this disclosure. First, as shown in Figure 1, in the text-to-image program PD1, the text-to-image module 100 generates a base image IM1 based on the text prompt PT. For example, when the text prompt PT, "beautiful sunset," is received from the user, the text-to-image module 100 will generate a landscape image with a sunset using artificial intelligence image generation technology.

[0021] Next, please refer to Figures 3 and 5. Figure 5 illustrates the image expansion drawing procedure PD2. In the image expansion drawing procedure PD2, the image expansion drawing module 200 obtains an expanded image IM8 based on the base image IM1. The preprocessing unit 220 of the image expansion drawing module 200 uses several edge pixels PL of the base image IM1 to fill a ring-shaped block B2 surrounding the base image IM1 to obtain a preprocessed image IM2.

[0022] The ring-shaped block B2 includes several corner blocks B21, B23, B25, and B27, and several rectangular connecting blocks B22, B24, B26, and B28. Rectangular connecting block B22 connects corner blocks B21 and B23; rectangular connecting block B24 connects corner blocks B23 and B25; rectangular connecting block B26 connects corner blocks B25 and B27; and rectangular connecting block B28 connects corner blocks B27 and B21.

[0023] The edge pixels (PL) include several corner pixels (PL1, PL3, PL5, PL7) and several side pixels (PL2, PL4, PL6, PL8). Side pixel PL2 is located between corner pixels PL1 and PL3; side pixel PL4 is located between corner pixels PL3 and PL5; side pixel PL6 is located between corner pixels PL5 and PL7; and side pixel PL8 is located between corner pixels PL7 and PL1.

[0024] Corner block B21 is filled using corner pixels PL1 of the base image IM1; rectangular connecting block B22 is filled using side pixels PL2 of the base image IM1; corner block B23 is filled using corner pixels PL3 of the base image IM1; rectangular connecting block B24 is filled using side pixels PL4 of the base image IM1; corner block B25 is filled using corner pixels PL5 of the base image IM1; rectangular connecting block B26 is filled using side pixels PL6 of the base image IM1; corner block B27 is filled using corner pixels PL7 of the base image IM1; and rectangular connecting block B28 is filled using side pixels PL8 of the base image IM1. This filling action can improve the accuracy and speed of extended drawing. In one embodiment, the filling action can be omitted.

[0025] Next, as shown in Figures 3 and 5, the cutting unit 230 of the image extension drawing module 200 cuts the pre-processed image IM2 into several images to be drawn, IM31, IM32, IM33, and IM34. The number of these images to be drawn, IM31, IM32, IM33, and IM34, is four. Image IM31 includes the upper left corner of the base image IM1 and the upper left corner of the annular block B2; image IM32 includes the upper right corner of the base image IM1 and the upper right corner of the annular block B2; image IM33 includes the lower right corner of the base image IM1 and the lower right corner of the annular block B2; and image IM34 includes the lower left corner of the base image IM1 and the lower left corner of the annular block B2. The dimensions of these images to be drawn, IM31, IM32, IM33, and IM34, are substantially the same.

[0026] In one embodiment, the images to be drawn IM31, IM32, IM33, and IM34 comprise a base image IM1 of the same size. In another embodiment, the images to be drawn IM31, IM32, IM33, and IM34 may comprise a base image IM1 of different sizes.

[0027] Please refer to Figure 6, which illustrates the relationship between the preprocessed image IM2 and the images to be drawn, IM31, IM32, IM33, and IM34. These images to be drawn, IM31, IM32, IM33, and IM34, partially overlap at the edges to facilitate image stitching.

[0028] Next, as shown in Figures 3 and 5, the masking unit 240 of the image extension drawing module 200 obtains several masks MK41, MK42, MK43, and MK44 based on the images to be drawn IM31, IM32, IM33, and IM34. In each mask MK41, MK42, MK43, and MK44, the pixels corresponding to 0 (the diagonal area) do not need to be expanded, while the pixels corresponding to 1 (the blank area) do need to be expanded. That is, the base image IM1 corresponding to the pixels of 0 will be retained, and only the ring-shaped tile B2 corresponding to the pixels of 1 needs to be expanded.

[0029] Then, as shown in Figures 3 and 5, the expansion unit 250 of the image expansion drawing module 200 expands the content of the annular blocks B2 of the portions of the images IM31, IM32, IM33, and IM34 to be drawn, based on these masks MK41, MK42, MK43, and MK44, to obtain several drawn images IM51, IM52, IM53, and IM54. Since the expansion actions of these images IM31, IM32, IM33, and IM34 to be drawn are performed separately in a pipeline manner, they do not simultaneously occupy computing resources and memory resources.

[0030] In the step of expanding the content of the ring-shaped patch B2 of the portions of the images IM31, IM32, IM33, and IM34 to be drawn, the expansion is performed using positive prompts, without the need for negative prompts. In other words, researchers found that negative prompts are not very helpful for content expansion; therefore, omitting negative prompts and using only positive prompts for content expansion can improve the expansion speed while maintaining a certain level of accuracy.

[0031] Next, as shown in Figures 3 and 5, the noise reduction unit 260 of the image extension drawing module 200 performs noise reduction on the drawn images IM51, IM52, IM53, and IM54.

[0032] Then, as shown in Figures 3 and 5, the stitching unit 280 of the image extension drawing module 200 stitches together these drawn images IM51, IM52, IM53, and IM54 to obtain an extended image IM8. Since the images to be drawn IM31, IM32, IM33, and IM34 partially overlap at the edges, the drawn images IM51, IM52, IM53, and IM54 can be stitched together smoothly without any abrupt lines.

[0033] Next, as shown in Figures 1 and 3, in the resolution adjustment program PD3, the resolution adjustment module 300 adjusts the resolution of the extended image IM8 to obtain the image IM9 with adjusted resolution. In this program, the resolution of the extended image IM8 can be adjusted according to the resolution of the display.

[0034] Then, as shown in Figures 1 and 3, in the hidden watermark compositing program PD4, the hidden watermark compositing module 400 composites the adjusted resolution image IM9 with a hidden watermark WM to obtain the generated image IM10.

[0035] In the above embodiments, a pipelined approach can be used to execute the text-to-image program PD1, the image expansion and drawing program PD2, the resolution adjustment program PD3, and the hidden watermark synthesis program PD4. Multiple programs can be executed simultaneously, which can significantly speed up the processing speed.

[0036] Please refer to Figure 7, which illustrates one way in which the image generation method disclosed herein achieves zero copy. In the embodiment of Figure 7, the neural processor (NPU) executes a text-to-image program PD1, an image expansion and rendering program PD2, a resolution adjustment program PD3, and a hidden watermark synthesis program PD4. After the text-to-image program PD1 is executed, the intermediate calculation results are stored in the memory of the neural processor (NPU). The image expansion and rendering program PD2 directly retrieves the intermediate calculation results of the text-to-image program PD1 from the memory of the neural processor (NPU). After the image expansion and rendering program PD2 is executed, the intermediate calculation results are stored in the memory of the neural processor (NPU). The resolution adjustment program PD3 directly retrieves the intermediate calculation results of the image expansion and rendering program PD2 from the memory of the neural processor (NPU). After the resolution adjustment program PD3 is executed, the intermediate calculation results are stored in the memory of the neural processor (NPU). The resolution adjustment program PD3 directly retrieves the intermediate calculation results of the hidden watermark synthesis program PD4 from the memory of the neural processor (NPU).

[0037] In the embodiment shown in Figure 7, intermediate computation results do not need to be transmitted to the central processing unit (CPU) first. The neural processing unit (NPU) retrieves intermediate computation results directly from its own memory, which can significantly save data transfer time.

[0038] Please refer to Figure 8, which illustrates another way in which the image generation method disclosed herein achieves zero-copying. In the embodiment of Figure 8, the graphics processing unit (GPU) executes a text-to-image program PD1, an image extension drawing program PD2, a resolution adjustment program PD3, and a hidden watermark synthesis program PD4. After the text-to-image program PD1 is executed, the intermediate calculation results are stored in the GPU's memory. The image extension drawing program PD2 directly obtains the intermediate calculation results of the text-to-image program PD1 from the GPU's memory. After the image extension drawing program PD2 is executed, the intermediate calculation results are stored in the GPU's memory. The resolution adjustment program PD3 directly obtains the intermediate calculation results of the image extension drawing program PD2 from the GPU's memory. After the resolution adjustment program PD3 is executed, the intermediate calculation results are stored in the GPU's memory. The resolution adjustment program PD3 directly obtains the intermediate calculation results of the hidden watermark synthesis program PD4 from the GPU's memory.

[0039] In the embodiment shown in Figure 8, intermediate calculation results do not need to be transmitted to the central processing unit (CPU) first. The graphics processing unit (GPU) retrieves the intermediate calculation results directly from its own memory, which can significantly save data transfer time.

[0040] In summary, although this disclosure has been presented above with examples, it is not intended to limit the scope of this disclosure. Those skilled in the art to which this disclosure pertains can make various modifications and refinements without departing from the spirit and scope of this disclosure. Therefore, the scope of protection of this disclosure shall be determined by the appended claims.

[0041] 100: Text to Image Module 200: Image Extension Drawing Module 220: Preprocessing unit 230: Cutting unit 240: Masking Unit 250: Extended Unit 260: Noise Reduction Unit 280: Splicing unit 300: Resolution Adjustment Module 400: Hidden Watermark Synthesis Module 1000: Image Generation System B2: Circular Tile B21, B23, B25, B27: Corner tiles B22, B24, B26, B28: Rectangular connection blocks CPU: Central Processing Unit GPU: Graphics Processor IM1: Base Image IM2: Pre-processed image IM31, IM32, IM33, IM34: Images to be drawn IM51, IM52, IM53, IM54: Images already drawn IM8: Expanded image IM9: Adjusted resolution image IM10, IM101, IM102, IM103: Generated Images NPU: Neural Processor MK41, MK42, MK43, MK44: Masking PD1: Text-to-Image Program PD2: Image Extension Drawing Program PD3: Resolution Adjustment Procedure PD4: Hidden Watermark Synthesis Process PL1, PL3, PL5, PL7: corner pixels PL2, PL4, PL6, PL8: Side pixels PT, PT1, PT2, PT3: Text prompts T1, T2, T3, T4, T5, T6, T7: Time points WM: Hide watermark

Claims

1. An image generation method, comprising: Run a text-to-image program to generate a base image based on a text prompt; An image expansion drawing procedure is executed to obtain an expanded image based on a base image. The image expansion drawing procedure includes: performing a filling operation on a ring-shaped patch surrounding the base image using a plurality of edge pixels of the base image to obtain a pre-processed image, wherein the filling operation fills the same column, row, or block of the ring-shaped patch with identical pixel values; cutting the pre-processed image into a plurality of images to be drawn, each image to be drawn including a portion of the base image and a portion of the ring-shaped patch; obtaining a plurality of masks based on the images to be drawn; performing an expansion operation on the portion of the ring-shaped patch of each image to be drawn based on the masks to obtain a plurality of drawn images, wherein the expansion operation is different from the filling operation; and stitching the drawn images together to obtain the expanded image. A resolution adjustment procedure is executed to adjust the resolution of the expanded image to obtain an image with adjusted resolution; and a hidden watermark compositing procedure is executed to composit a hidden watermark onto the image with adjusted resolution to obtain a generated image; wherein the text-to-image procedure, the image expansion and rendering procedure, the resolution adjustment procedure, and the hidden watermark compositing procedure are executed in a pipeline manner through a graphics processing unit (GPU) and a neural network processing unit (NPU).

2. The image generation method as described in claim 1, wherein in the image expansion drawing procedure, the resolution adjustment procedure, and the hidden watermark synthesis procedure, an intermediate calculation result is obtained directly from one of the memory of the graphics processor or one of the memory of the neural processor.

3. The image generation method as described in claim 1, wherein the annular tile includes a plurality of corner tiles and a plurality of rectangular connecting tiles, the edge pixels include a plurality of corner pixels and a plurality of side pixels, the rectangular connecting tiles connect the corner tiles, the corner tiles are filled with the corner pixels of the base image, and the rectangular connecting tiles are filled with the side pixels of the base image.

4. The image generation method as described in claim 1, wherein the images to be drawn are substantially the same size.

5. The image generation method as described in claim 1, wherein the images to be drawn comprise the base image of portions of the same size.

6. The image generation method as described in claim 1, wherein the number of images to be drawn is 4.

7. The image generation method as described in claim 1, wherein the images to be drawn partially overlap.

8. The image generation method as described in claim 1, wherein in each of the masks, pixels corresponding to 0 do not require content expansion, and pixels corresponding to 1 require content expansion.

9. An image generation system, comprising: A text-to-image module is used to execute a text-to-image program to generate a base image based on a text prompt. An image expansion drawing module is used to execute an image expansion drawing program to obtain an expanded image based on a base image. The image expansion drawing module includes: a preprocessing unit, which fills a ring-shaped patch surrounding the base image with a plurality of edge pixels of the base image to obtain a preprocessed image, wherein the filling operation fills the same column, row, or block of the ring-shaped patch with identical pixel values; a cutting unit, which cuts the preprocessed image into a plurality of images to be drawn, each image to be drawn including a portion of the base image and a portion of the ring-shaped patch; a masking unit, which obtains a plurality of masks based on the images to be drawn; an expansion unit, which expands a portion of the ring-shaped patch of each image to be drawn based on the masks to obtain a plurality of drawn images, wherein the expansion operation is different from the filling operation; and a stitching unit, which stitches the drawn images to obtain the expanded image. A resolution adjustment module is used to execute a resolution adjustment program to adjust the resolution of the expanded image and obtain an image with adjusted resolution; and a hidden watermark compositing module is used to execute a hidden watermark compositing program to composit a hidden watermark on the image with adjusted resolution and obtain a generated image; wherein the text-to-image program, the image expansion and rendering program, the resolution adjustment program and the hidden watermark compositing program are executed in a pipeline manner through a graphics processing unit (GPU) and a neural network processing unit (NPU).

10. The image generation system as claimed in claim 9, wherein in the image expansion drawing process, the resolution adjustment process, and the hidden watermark synthesis process, an intermediate computation result is obtained directly from a memory of the graphics processor or a memory of the neural processor.

11. The image generation system of claim 9, wherein the annular tile includes a plurality of corner tiles and a plurality of rectangular connecting tiles, the edge pixels include a plurality of corner pixels and a plurality of side pixels, the rectangular connecting tiles connect the corner tiles, the corner tiles are filled with the corner pixels of the base image, and the rectangular connecting tiles are filled with the side pixels of the base image.

12. The image generation system as described in claim 9, wherein the images to be drawn are substantially the same size.

13. The image generation system as described in claim 9, wherein the images to be drawn comprise the base image of portions of the same size.

14. An image generation method, comprising: Run a text-to-image program to generate a base image based on a text prompt; An image expansion drawing procedure is executed to obtain an expanded image based on a base image. The image expansion drawing procedure includes: performing a filling operation on a ring-shaped patch surrounding the base image using a plurality of edge pixels of the base image to obtain a pre-processed image, wherein the filling operation fills the same column, row, or block of the ring-shaped patch with identical pixel values; cutting the pre-processed image into a plurality of images to be drawn, each image to be drawn including a portion of the base image and a portion of the ring-shaped patch; obtaining a plurality of masks based on the images to be drawn; performing an expansion operation on the portion of the ring-shaped patch of each image to be drawn based on the masks to obtain a plurality of drawn images, wherein the expansion operation is different from the filling operation; and stitching the drawn images together to obtain the expanded image. A resolution adjustment procedure is executed to adjust the resolution of the expanded image to obtain an image with adjusted resolution; and a hidden watermark compositing procedure is executed to composite a hidden watermark on the image with adjusted resolution to obtain a generated image; wherein in the image expansion drawing procedure, the resolution adjustment procedure, and the hidden watermark compositing procedure, an intermediate computation result is obtained directly from a memory of a graphics processing unit (GPU) or a memory of a neural network processing unit (NPU).

Images