A

SEMINAR REPORT ON

“Creating An Innovative Image Generator With

Open Ai Text Prompt”
Submitted To

SAVITRIBAI PHULE PUNE UNIVERSITY

In Fulfillment of the Requirement for the Awarded of

BACHELOR OF ENGINEERING ININFORMATION

TECHNOLOGY
BY
Mr. Pagar Yogesh Y.[72244894f]
Under The Guidance of
Prof. Narode S. S.

DEPARTMENT OF INFORMATION TECHNOLOGY

S.N.D. COLLEGE OF ENGINEERING & RESEARCH
YEOLA – 423401
2024 – 2025.
Creating An Innovative Image Generator With Open Ai Text Prompt
S.N.D. COLLEGE OF ENGINEERING & RESEARCH
CENTRE, BABHULGAON
YEOLA – 423401
2024 – 2025

CERTIFICATE
This is certified that the project entitled

“Creating An Innovative Image Generator With

Open Ai Text Prompt”

Submitted By

Mr. Pagar Yogesh Y.[72244894F]

Is a bonafide work carried out by Students under the supervision of

Prof.NarodeS. S.and it is submitted towards the fulfillment of the
requirement of Bachelor ofEngineering (Information Technology).

Prof. Narode S. S. Prof. Wadghule Y. M. Dr. Rokade P.P. Dr. Yadav D.M

Project Guide Project Co-ordinator Head Of Dept Principal

Creating An Innovative Image Generator With Open Ai Text Prompt

ABSTRACT

This conversion of textual descriptions into visual representations offers great creative
opportunities as well as responds to the burgeoning demand for compelling visual content in
our increasingly image-driven society. The field evolved from basic system-level machine
learning technology to sophisticated deep learning models capable of generating images from
text prompts. This paper recaps the major developments in text-to-image synthesis, focusing
on model and technique evolution, including but not limited to GANs, and more recently
introduced diffusion models that have surpassed others.

Text-to-image synthesis traces its history from the earliest, primitive systems
to the current state-of-the-art deep learning-based tools. The capability was relatively limited
in these systems, and it has completely changed with complex neural networks that could
understand and interpret textual input much more efficiently. GANs have been considered a
backbone for generations in the generation of images. This works through a dual-network
structure that comprises a generator and discriminator inorder to produce highly realistic
images. Its strengths are there, but mode collapse and inability to generate high-resolution
images are some of the drawbacks of this approach. Diffusion models, in recent times, have
emerged as much more competitive alternatives that present stunning capabilities of
producing high-fidelity images through iterative refinement processes.

Although very interesting, lots of challenges still do exist in text-to-image

synthesis space. Probably the biggest challenge is semantic coherence between text and
images, more particularly when dealing with ambiguous or complex prompts. There are ethics
involved in the final synthesized images, especially considering biased or distorted
representations due to biases in training data. Their performance is assessed using various
metrics like FID and IS, while also human evaluation becomes applicable. Each one of these
metrics reveals implications for developing the model while the researchers refine their
approaches The quality of performance largely depends on the primary datasets used in
training text-to-image models. The differences in size, diversity, or composition may
decisively impact the quality of images synthesized. It synopsizes the state-of-the-art on text-
to-image synthesis: key findings and directions forward. This kind of analysis makes
importance on creativity, accessibility, and ethical consideration of the implications of these
technologies in the context of responsible innovation within this rapidly changing field.
Creating An Innovative Image Generator With Open Ai Text Prompt

ACKNOWLEDGEMENT

We take this opportunity to acknowledge all the people who have helped us whole heartedly
in every stage of this seminar. We are indebtedly grateful to the Head of Information Technology
Department, Dr. Rokade P. P. for his valuable support. We would like to express our deep-felt
gratitude to our Project Guide, Prof. Narode S. S. for giving us an opportunity to work and for
their advice, encouragement, and constant support. We wish to thank them for extending us the
greatest freedom in deciding the direction and scope of our seminar. It has been both a privilege
and a rewarding experience working with him/her. We also extend our sincere thanks to Principal
of S.N.D. COE & RC Dr. Yadav D.M. for their valuable inspiration. We would also like to
thank our classmates here at S.N.D.COE & RC, for all the wonderful times we have had with
them. Their
valuable comments and suggestions have been vital to the completion of this work. We want
to thank thefaculty of S.N.D.COE & RC and the staff for providing us the means to complete
our diploma.
And finally, we are grateful to our parents and siblings for their love, understanding,
encouragement and support.

Mr. Pagar Yogesh Y.

Creating An Innovative Image Generator With Open Ai Text Prompt

PAGE INDEX

Sr .No Topic Page no

1 INTRODUCTION 7

2 LITERATURE SURVAY 8

3 MOTIVATION 9

4 PURPOSE & SCOPE 10

5 OBJECTIVES 11

6 METHODS & ALGORITHM 12-13

7 WORKING 14-18

8 APPLICATIONS 19-20

9 FUTURE SCOPE 21

10 ADVANTAGES & 22-23

DISADVANTAGES

11 CONCLUSION 24

12 REFERENCES 25
Creating An Innovative Image Generator With Open Ai Text Prompt

FIGURE INDEX

Figure no. Figure Page No.

1 TEXT TO IMAGE GENERATOR 9

2 ALGORITHM DID 12

3 WORKING OF TEXT -IMAGE PROMPT 17

Creating An Innovative Image Generator With Open Ai Text Prompt

INTRODUCTION

The speed at which Artificial Intelligence has been improving across all applications is
just mind-blowing, particularly in the text-to-image generation space. This technology unifies
natural language and computer vision to generate images out of text descriptions. AI techniques
are capable of rendering visuals in forms such as vector graphics, 3D renders, and photorealistic
images by interpreting the given text as a set of instructions. Systemically Understanding the
Relationship between Vision and Language: The milestone approach toward achieving human-
like intelligence has developed systems that understand the intricate relationship between vision
and language. Recent breakthroughs in deep learning have shown that there is much to be gained
with new methods and applications in processing images in computer vision.

This approach focuses on discovering deep, hierarchical models that effectively represent
probability distributions of diverse data types used in AI systems. Within this framework, image
synthesis, of course, plays the most critical role while generating completely new images and
modifying existing ones. The scope of its applications encompasses a wide range of tasks,
including image editing, art generation, computer-aided design, and virtual reality, among others.
In addition, the capacity for AI to generate imagery from text opens exciting possibilities in the
creative industry, which enhances artistic expression and simplifies work flows. With these
technologies evolving further, they will be constrained only to change the way we communicate
digitally, and image synthesis, in this case, turns out not only as a technical achievement but also
an unbridled source of creativity and innovation in real-world applications.

The implications of this technology go beyond aesthetics, offering new paths for
storytelling, communication, even education remaking our understanding about what visual
media may become in the digital era.

7
Creating An Innovative Image Generator With Open Ai Text Prompt

LITERATURE SURVEY

Study/Source Objective Methodology Key Findings Relevance to

Project
DALL-E: Explore capabilities Deep learning, High-quality, Foundation for
Creating Images of DALL-E in GANs, CLIP model diverse image text-to-image
from Text generating images integration generation; generation
(OpenAI) from text. understanding of
text-to-image
relationships.
Text-to-Image Create images from Using a two-stage Successful in Relevant for user
Synthesis (Reed detailed textual generation process. generating complex input handling.
et al., 2016) descriptions. images from
detailed text.
Ethical Examine ethical Qualitative analysis Raises concerns Important for
Considerations implications of AI- of AI impact on about copyright, addressing ethical
in AI Art generated art. creativity. originality, and bias issues.
(Elgammalet al., in generated art.
2017)
Image Synthesis Improve image Conditional GANs; Improved visual Insights for
from Text fidelity and data augmentation. coherence; context- enhancing image
(Zhang et al., relevance to driven image quality.
2018) prompts. outputs.
Human-AI . Explore Case studies of AI as a creative Guides user
Collaboration in collaborative roles human-AI partner; enhances interaction design.
Art Creation of AI in creative collaborations. human creativity.
(McCormack et processes.
al., 2019)
User Experience Investigate user Surveys and user User preferences Critical for
in AI-generated satisfaction with testing. for control and designing user
Content (Liu et AI-generated customization; interaction.
al., 2020) content. importance of
feedback
mechanisms.

8
Creating An Innovative Image Generator With Open Ai Text Prompt

Motivation

In fact, it is the primary drive behind a revolutionary image generator based on OpenAI's text
prompts: creativity and technology intersect. During times where everyone is running so fast for their
life in the virtual world, visual content is important for communication, marketing, and storytelling.
However, not everyone has the artistic talent to bring those ideas to life. Using the latest state-of-the-
art machine learning algorithms, any person, regardless of experience in art, can communicate their
thoughts and imagination effectively with high-quality images. The rapid development of AI also
offers an exciting opportunity to develop novel forms of artistic expression that can be discussed
across educational, entertainment, and other forums.

9
Creating An Innovative Image Generator With Open Ai Text Prompt

PURPOSE & SCOPE

Purpose of the Project

The principal goal of this project is to design a picture-generation tool that can be user-friendly
enough to change textual descriptions into powerful images. This assistant tool will help users come
up with images of their ideas, concepts, or stories. Through this development and adding feedback
mechanisms together with strategies for continuous improvement, the image generation process will
be refined so that it produces high-quality outputs in response to user expectations. Ultimately, it
hopes to improve the ability of the individual, along with cultivating a community that begins with
the creative process.

Scope

This project is full of key areas within scope:

User Interface Design: Ensuring an intuitive interface that can accommodate user-in prompts, and
then potentially proceeds to view generated images.

Backend Development: Set up robust API integration for OpenAI's image generation capabilities
besides creating a database to store user information and feedback.

Image Generation Logic. Effective algorithms in terms of prompt processing and error handling
have to be applied to ensure relevant and quality outputs in images.

Quality Control. The feedback mechanism and continuous improvement processes must also have
recourse to refine the generator from the user's insight.

Testing and Validation. Extensive testing phases and validation occur to ensure performance,
usability, and reliability under various conditions.

Launch and Community Engagement: The planning of the application launch, along with building
a community around it to interact with, share, and collaborate.

10
Creating An Innovative Image Generator With Open Ai Text Prompt

OBJECTIVES

1) Examine AI Interpretation: Explain how the models in OpenAI work to take textual
descriptions to their equivalent visual representation, with an underpinning of mechanisms.

2) Design User-Friendly Interfaces: Make an intuitive and accessible interface for users to input
textual prompts and get generated images easily.

3) Detail Technical Architecture: Describe how OpenAI's ability to generate images could
correctly be integrated into a coherent, working application through technical requirements and
architecture.

4) To determine Evaluation Criteria: Define quality, relevance, and coherence explicit criteria for
the images produced with respect to the user-provided prompts.

5) To encourage User Creativity: Explore how this generator can stimulate user creativity by
producing imagination and diversity visuals inspired by their textual inputs.

6) To assess practical applications: Explore its practical applications in real-life fields such as art,
marketing, and education.

11
Creating An Innovative Image Generator With Open Ai Text Prompt

TEXT-TO-IMAGE GENERATION METHODS

This section provides an overview of relevant studies on text-to-image generative models.

Due to the diversity of the generative models and the vast amount of associated literature, this
study narrows its focus to the two cutting hedge types of deep learning generative models: GANs
and diffusion models. A. TEXT-TO-IMAGE GENERATION USING GANS Since its
introduction in 2014, GAN-based text-to-image synthesis has been the subject of numerous
studies, leading to significant advancements in the field. Reed et al. [42], working upon the
foundation laid by deep convolutional GANs [43], were the first to investigate the GAN-based
text-to-image synthesis technique. Earlier models could create images based on universal
constraints like a class label or caption, but not pose or location. Therefore, the Generative
Adversarial What-Where Network (GAWWN) [44] was proposed, which is a network that
generates images based on directions about what to draw and where to draw it. It demonstrates
the ability to generate images based on free-form text descriptions and the precise location of
objects. GAWWN enables precise location management through the use of a bounding box or a
collection of key points. Stacked Generative Adversarial Networks (Stack GAN) [45] established
a two-stage conditioning augmentation approach to boost the diversity of synthesized images and
stabilize conditional-GAN training. Using the provided text description as input, the Stage-I GAN
generates lower solution images of the initial shape and colors of the object. High-resolution (e.g.,
256 × 256) images with photorealistic features are generated by the Stage-II GAN using the
results from Stage-I and the descriptive text.

Fig1. Text To Image Generate Method

12
Creating An Innovative Image Generator With Open Ai Text Prompt

Algorithm
1. Input Prompt:
• Receive a textual description from the user, detailing the desired image (e.g., "A sunset over
a mountain range").

2. Preprocessing:
• Tokenize the input text to convert it into a format suitable for the model. This may
involve:
• Converting the text into tokens using a tokenizer (e.g., WordPiece or Byte Pair
Encoding).
• Creating embeddings for the tokens using a pre-trained embedding model.
3. Text Encoding:
• Pass the tokenized input through the DALL·E text encoder to obtain a high-dimensional text
representation (embedding).
• Ensure that the embedding captures the semantics and context of the original prompt.
4. Image Generation Using DALL·E:
• Feed the text embedding into DALL·E's decoder to generate an initial image.
• The model utilizes a trained transformer architecture to synthesize an image that aligns with
the provided text prompt.
5. Post-Processing:
• Apply any necessary post-processing steps to the generated image, such as:
• Upscaling the image resolution using techniques like super-resolution.
• Enhancing the image quality through filtering or noise reduction.
6. Quality Assessment (if using GANs):
• If using a GAN model for additional refinement, input the generated image into a
discriminator network.
• The GAN’s generator can then be used to improve the image based on feedback from the
discriminator to ensure realism and coherence.
7. Output Image:
• Return the final generated image to the user.
• Optionally, provide options for the user to regenerate or refine the image based on further
prompts or modifications.
8. User Interaction:
• Allow the user to provide feedback or additional prompts to iteratively refine the image.
• Implement a system to log user interactions for continuous improvement of the model.

FIG.2.Algorithm Of Ai Creation

13
Creating An Innovative Image Generator With Open Ai Text Prompt

working
Creating an innovative image generator utilizing OpenAI’s text prompts entails a
comprehensive, multi-faceted approach that integrates technical development, user experience
design, and effective implementation strategies. Below is a detailed overview of the process
involved in this project.

1. Project Planning

A. Define Goals

The initial phase involves establishing clear objectives for the image generator, which may
encompass the creation of unique artistic styles and ensuring accessibility for a diverse user base.
Additionally, it is imperative to identify specific use cases that the generator will serve, such as
generating marketing visuals, facilitating art creation, or producing educational content tailored
to various audiences.

B. Assemble a Team

A diverse project team is essential for success. This team should comprise professionals with
varying expertise, including software developers, UI/UX designers, and data scientists, all of
whom will collaborate to ensure the project’s technical and creative dimensions are effectively
addressed.

2. Research and Feasibility Study

A. Market Analysis

Conducting a thorough market analysis is crucial to understanding the competitive landscape of

existing image generators, such as DALL-E and Midjourney. This analysis should focus on
evaluating their strengths and weaknesses, thereby identifying gaps in the market that your
generator can uniquely address, such as enhanced customization options or improved user
accessibility features.

14
Creating An Innovative Image Generator With Open Ai Text Prompt

B. Technical Feasibility

A careful review of OpenAI’s API documentation and terms of use is necessary to ascertain the
integration capabilities of the image generator. Furthermore, assessing the technical requirements,
including server capabilities and storage needs, will provide insights into the infrastructure
necessary for supporting the application’s operational demands.

3. Design Phase

A. User Interface (UI) Design

The design phase commences with the creation of wireframes that outline the application’s layout
and user flow. Following this, the development of interactive prototypes is essential to visualize
user interactions and to gather feedback, thereby facilitating iterative improvements prior to full-
scale implementation. This phase ensures that the user interface is not only aesthetically pleasing
but also intuitive, enhancing overall user experience.

B. Prompt Input Mechanism

• Design a text input field with suggestions or examples to guide users.

• Consider adding a prompt history feature for easy reuse of past prompts.

4. Development

A. Backend Development

• API Integration: Set up communication with OpenAI’s image generation API, ensuring secure
and efficient data exchange.
• Image Processing: Implement backend logic to handle prompt processing, manage image
generation requests, and store generated images in the database.
• Database Setup: Create a database to store user data, generated images, and user feedback,
facilitating future enhancements and user engagement.

B. Frontend Development

• Dynamic Content Rendering: Utilize JavaScript frameworks (e.g., React, Vue.js) to

dynamically display generated images, enhancing user engagement.
15
Creating An Innovative Image Generator With Open Ai Text Prompt

• User Interaction Features: Implement features such as customization options, sliders for
adjustments, and buttons for sharing images, allowing users to personalize their experience.

5. Image Generation Logic

A. Prompt Processing

• Develop algorithms to interpret user prompts effectively, employing natural language

processing (NLP) techniques to parse and understand nuances in user requests.

B. Image Generation

• Use OpenAI’s API to generate images based on the processed prompts, ensuring alignment
with user expectations.
• Implement robust error handling to manage cases where the API may not return expected
results, providing users with helpful feedback or alternative options.

6. Quality Control

A. User Feedback Mechanism

• Implement a feedback system where users can rate the generated images and provide
comments.
• Utilize this feedback to refine prompt processing and image generation algorithms, ensuring
continuous improvement.

B. Continuous Improvement

• Analyze user feedback and image quality metrics regularly to make iterative enhancements,
addressing any recurring issues or areas for development.

7. Testing and Validation

A. Alpha and Beta Testing

• Conduct alpha testing within the development team to identify and resolve bugs before wider
release.
• Roll out a beta version to a select group of users for real-world testing, gathering insights on
usability and performance.

8. B. Performance Testing
16
Creating An Innovative Image Generator With Open Ai Text Prompt

• Test the application under various load conditions to ensure it can efficiently handle multiple
users and requests simultaneously.

9. Launch and Marketing

A. Launch Strategy

• Plan a launch event or marketing campaign to generate excitement around the image
generator.
• Leverage social media, influencer partnerships, and content marketing strategies to effectively
reach the target audience.

B. Community Building

• Create forums or social media groups for users to share their experiences and artwork,
fostering a sense of community.
• Encourage user-generated content by hosting challenges or contests, further engaging users
and promoting creativity.

10. Post-Launch Support

A. Ongoing Maintenance

• Set up a system for regular updates and maintenance of the application.

• Monitor server performance and user activity to ensure a smooth experience.

B. User Support

• Provide robust customer support through FAQs, tutorials, and direct assistance channels.

10. Ethical Considerations

A. Content Moderation

• Develop and implement guidelines for acceptable content generation.

• Use moderation tools to prevent the generation of harmful or inappropriate images.

17
Creating An Innovative Image Generator With Open Ai Text Prompt

18
Creating An Innovative Image Generator With Open Ai Text Prompt

THE APPLICATIONS

There are many uses for developing a creative image generator with OpenAI's text prompts in a
variety of industries. These are a few thorough applications:
1. The Arts and Design
• Digital Art Creation: By offering detailed instructions, artists can produce original works
of art, whether they are completed pieces or just inspiration.
• Concept Art: Using descriptions of characters or environments, designers may quickly bring
ideas to life in video games or movies.
• Graphic Design: To convey particular themes or messages, marketers can produce images for
campaigns or social media postings.

2. Entertainment • Storyboarding: Filmmakers can better envisage scenes prior to production

by creating visual storyboards based on scripts.
The process of creating video game assets can be streamlined by allowing developers to
generate character models, environments, or objects based on textual descriptions.

3. Education: Teachers can produce illustrative learning materials.

• Visual Aids: Producing visuals to go along with written material helps improve
comprehension of disciplines like history and physics.

4. Promotion and Promotion

• Custom Ad Creatives: Based on insights about the target audience or brand messaging,
marketers can create graphics specifically suited for campaigns.
• Social Media Content: Produce captivating images quickly to keep up an active social media
presence while adhering to popular subjects.

5. Medical Care • Medical Illustration: Producing visuals for instructional reasons, including
depicting medical processes or conditions.

Patient Education: Creating visuals to help patients understand diagnoses or treatment options
more clearly.
6. Fashion • Design Prototyping: Fashion designers can use descriptions to visualize clothing
concepts, which facilitates the ideation process.
• Trend Visualization: Producing visuals for marketing and design direction that represent
prevailing trends or styles.

19
Creating An Innovative Image Generator With Open Ai Text Prompt

7. Customization and Personalization

• Custom Gifts: Depending on the hobbies and preferences of loved ones, users can design
personalized gifts (such as prints or cards).
Home Decor: Generating custom artwork for interior design, tailored to a specific aesthetic or
color scheme.

8. Cultural Preservation
• Historical Reconstructions: Creating visuals based on descriptions of historical events,
places, or figures, aiding in education and preservation efforts.
9. Accessibility
• Visualizing Descriptions for the Visually Impaired: Generating images based on
detailed verbal descriptions can help make visual content more accessible.
10. Research and Development
• Rapid Prototyping: In fields like architecture or product design, generating images
based on textual specifications can speed up the ideation phase.

20
Creating An Innovative Image Generator With Open Ai Text Prompt

FUTURE SCOPE
An inventive image generator driven by OpenAI's text prompts has a wide range of
constantly developing potential uses. As technology develops, it will improve user engagement,
open up new creative processes, and provide innovative solutions for a variety of businesses.
But it will also necessitate constant attention to user impact and ethical issues.

Personalized Art Creation: Using text prompts, designers and artists can create original
artwork that is suited to particular themes or feelings.

Rapid Prototyping: The ability of designers to produce graphic concepts quickly aids in
streamlining the creative process.

Dynamic game content improves replay ability and immersion by allowing developers to
produce assets that alter in response to player input or story developments.

Storyboarding and Concept Art: Project timescales can be accelerated by producing images
for pitches or story development quickly.

Personalized Campaign Visuals:

To increase interaction, marketers might produce customized images that closely match
campaign messaging or target audiences.

Visual Aids for Learning: To accommodate different learning styles, educators might produce
personalized pictures or infographics to clarify difficult ideas.

Simulation and Virtual Training: For training in emergency services, the military, or medical,
realistic situations can be created.

material Generation: Without investing a lot of effort, influencers and producers can swiftly
produce images for posts, increasing the diversity of material.

Brand Consistency: Businesses can maintain brand aesthetics by generating images that fit
predefined styles.

21
Creating An Innovative Image Generator With Open Ai Text Prompt

ADVANTAGES & DISADVANTAGES

➢ Advantages
1. Creativity Enhancement:
• Diverse Output: Image generators can produce a wide range of images from a single prompt,
fostering creativity and inspiration.
• Novelty: They can create unique images that may not have been envisioned by human artists,
leading to innovative concepts.
2. Accessibility:
• User-Friendly: Non-artists can create professional-quality images without needing advanced
design skills.
• Democratization of Art: More people can engage in creative processes, reducing barriers to
entry.
3. Speed and Efficiency:
• Rapid Prototyping: Users can generate multiple visual ideas quickly, which is valuable in fields
like marketing and product design.
• Instant Revisions: Immediate adjustments based on user feedback can streamline the creative
process.
4. Customization:
• Tailored Outputs: Users can customize prompts to refine results, leading to more specific
imagery that meets their needs.
• Iterative Design: Continuous prompting allows for evolution and improvement of designs over
time.
5. Cost-Effectiveness:
• Reduced Labor Costs: Businesses can lower costs by using AI for initial image drafts instead of
hiring multiple designers.
• Scalable Solutions: One AI can cater to numerous projects simultaneously.

22
Creating An Innovative Image Generator With Open Ai Text Prompt

➢ Disadvantages

1. Quality Control:

o Inconsistent Outputs: Generated images may not always meet quality standards or
expectations, requiring further refinement.
o Potential for Errors: AI can misinterpret prompts, leading to irrelevant or undesirable
images.
2. Creativity Limitations:
o Dependency on Prompts: The quality of the output is heavily dependent on the specificity
and creativity of the input prompts.
o Homogenization of Ideas: Over-reliance on AI-generated images might lead to a lack of
diversity in artistic expression.
3. Ethical Concerns:
o Copyright Issues: Generated images may unintentionally mimic existing works, raising
questions about ownership and copyright infringement.
o Deepfake Risks: The technology can be misused to create misleading or harmful imagery.
4. Technical Challenges:
o Resource Intensive: High-quality image generation requires significant computational
resources, which may not be accessible to everyone.
o Learning Curve: Users may need to learn how to craft effective prompts to get the desired
results.
5. Emotional Disconnect:
o Lack of Human Touch: AI-generated art may lack the emotional depth and personal
connection often found in human-created works.

23
Creating An Innovative Image Generator With Open Ai Text Prompt

CONCLUSION

Indeed, the creation of a new tool from OpenAI's text prompts is one step for giant leaps
in creative technology. As we make use of advanced natural language processing and machine
learning algorithms in handling the transformation of descriptive texts into visually interesting
images, we can clearly take a great step forward with such tools. This, in turn, opens up vistas for
artistic expression, design, and storytelling and democratizes, at least with respect to the quality
of images that can be achieved, the creative process even for those lacking experience. It also
enhances collaboration between disciplines: it sparks creativity and fosters innovation.

24
Creating An Innovative Image Generator With Open Ai Text Prompt

REFEENCES
Electronics for you & information technology magazines.
IEEE microwaves magazine
www.wikipedia.org

OPEN AI-DALL-E

[1] M. Ding, Z. Yang, W. Hong, W. Zheng, C. Zhou, D. Yin, J. Lin, X. Zou, Z. Shao, H. Yang,
and J. Tang, ‘‘Cog View: Mastering text-to-image generation via transformers,’’ in Proc. Adv.
Neural Inf. Process. Syst., vol. 24, May 2021, pp. 19822–19835.
[2] M. Ding, W. Zheng, W. Hong, and J. Tang, ‘‘CogView2: Faster and better text-to-image
generation via hierarchical transformers,’’ 2022, arXiv:2204.14217v2.
[3] S. Gu, D. Chen, J. Bao, F. Wen, B. Zhang, D. Chen, L. Yuan, and B. Guo, ‘‘Vector quantized
diffusion model for text-to-image synthesis,’’ in Proc. IEEE/CVF Conf. Comput. Vis. Pattern
Recognit. (CVPR), Nov. 2021, pp. 10686–10696.
[4] Saleema Amershi, Kori Inkpen, Jaime Teevan, Ruth Kikin-Gil, Eric Horvitz, Dan Weld,
Mihaela Vorvoreanu, Adam Fourney, Besmira Nushi, Penny Collisson, Jina Suh, Shamsi Iqbal,
and Paul Bennett. 2019. Guidelines for Human-AI Interaction. 1–13
[5] Autodesk. 2022. Autodesk Screencast. https://knowledge.autodesk.com/
community/screencast Retrieved September 15, 2022.
[6] Marcelo Bernal, John R. Haymaker, and Charles Eastman. 2015. On the role of computational
support for designers in action. Design Studies 41 (2015), 163–182.
[7] Gwern Branwen. 2020. Gpt-3 creative fiction. https://www.gwern.net/GPT-3
[8] Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla
Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal,
Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M.
Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz
Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec
Radford, Ilya Sutskever, and Dario Amodei. 2020. Language Models are Few-Shot Learners.

25