-
Prototype of add-on interface to interact with the aws api.
-
Example of output from AWI Lambda function when inputted an image.
-
-
-
-
-
-
-
-
-
Inspiration
The inspiration for Post Pals came from the desire to streamline and enhance the creative process for designers. We noticed that finding the right fonts, colors, and complementary images can be time-consuming, especially for those working on multiple projects or tight deadlines. We wanted to create a tool that could simplify these tasks by leveraging AI to deliver personalized, high-quality design resources in real time. The goal was to empower users to be more creative while saving valuable time, allowing them to focus on their core ideas.
What it does
Post Pals is an Adobe Express add-on that uses AI to analyze images and offer a range of resources to enhance your designs. By inputting an image, the tool extracts dominant colors, suggests matching fonts, and provides relevant clip art, making your designs more cohesive and visually engaging. It’s designed to serve both workplace and personal projects, helping users elevate their visual creations by making the right design choices faster and easier.
How we built it
We started by defining the architecture using AWS Lambda and S3 to process user-uploaded images. The images are analyzed by a Lambda function, which uses Pillow for image processing and KMeans clustering (via scikit-learn) to extract dominant colors. These colors are then used to suggest appropriate font themes. The infrastructure was automated with Terraform, while MongoDB served as the database to store image-related data. We then developed a front-end interface using HTML and JavaScript, integrating it with the Adobe Express platform through APIs to provide a seamless user experience.
Challenges we ran into
We encountered several technical challenges, including configuring AWS Lambda permissions and overcoming packaging issues with its dependencies. Ensuring that the correct libraries were included in the deployment package was a significant hurdle. Additionally, debugging permission errors to allow Lambda to properly access S3 resources required several iterations. Another challenge was optimizing the function to handle a wide range of image formats and sizes while maintaining accuracy in color extraction and font suggestion.
Accomplishments that we're proud of
Although we did not build a fully functional Adobe express add on, we successfully created working API's and a website two vital steps to creating the add on.
What we learned
Throughout this project, we learned a great deal about cloud computing, particularly the nuances of AWS Lambda and serverless architectures. We deepened our knowledge of image processing techniques and machine learning algorithms like KMeans clustering. Additionally, working on an integration for Adobe Express provided valuable insights into API management and how to enhance user experience in creative platforms.
What's next for Postpals
Next, we plan to expand Post Pals by adding more advanced features such as customizable font and color options, giving users even more control over their design outputs. We also aim to integrate additional design platforms beyond Adobe Express, making Post Pals a go-to tool for creatives across various industries. Furthermore, we will continue to refine the AI algorithms to offer even more personalized and relevant suggestions, helping users create with ease and confidence.
Log in or sign up for Devpost to join the conversation.