5 Image APIs You Can Use For Your Next Project in 2022
Images are essential in this digital world. No matter whether you are running an e-commerce business, managing a social media account, building an app, or running a blog, you will need to use images to catch the attention of your audiences. Sometimes, acquiring or producing images for your project could be troublesome due to copyright issues, the amount of time and effort needed, and the lack of creativity.
Luckily, there are many technologies and tools to help you generate images for different purposes, such as using image APIs. These APIs are especially useful when it comes to saving time in producing visuals and automating processes. Here are 5 image APIs you can use for different purposes to make your work easier:
1. Unsplash Image API
The Unsplash Image API provides unlimited API requests for developers to add images to their apps without any speed or quota limits. The API request returns a dynamic image URL powered by Imgix, a powerful image manipulation service for dynamic image URLs for every request. This means that the image can easily be manipulated to adjust the dimensions and quality in real-time by simply adjusting the query parameters of the image URL. This enables image transformation in real-time client-side, without any API calls.
2. Generated Photos
Generated Photos is an AI image generator to generate unique, worry-free model photos so that the faces on the photos can be used for any purpose without worrying about copyright, distribution rights, infringement claim, or royalties. They can be integrated easily into any app via their REST API for free for a limited usage for personal, non-commercial uses while the commercial license is available starting at $0.006/call. For free accounts, the Generated Photos API currently places a limit of 500 requests per month and the rate limit can be increased up to 100,000 requests per month with a paid subscription.
The API currently has only one endpoint:
GET /v1/faces. Optional parameters like “confidence”, “emotion”, “gender”, “age”, “ethnicity”, etc. can be inserted to generate faces with different features and emotions like the example below.
The image below shows a screenshot of a “joyful white young-adult female with long brown hair and brown eyes” generated using Generated Photos on their website:
3. DALL·E/DALL·E Mini
In the year 2021, OpenAI, an AI research and deployment company introduced DALL·E, an AI system that generates images from textual descriptions. It is a 12-billion parameter version of GBT-3 (an autoregressive language model that uses deep learning to produce human-like text) trained to generate images from text descriptions, using a dataset of text-image pairs. It does not only generate an image from scratch but can also generate any rectangular region of an existing image that extends to the bottom-right corner, in a way that is consistent with the text prompt.
The DALL·E neural network is trained to create images from text captions for a wide range of concepts expressible in natural language. For example, a user can type “an illustration of a baby daikon radish in a tutu walking a dog” and a series of AI-generated images which accurately illustrate the description will be displayed. This is one of the examples that is shown on the official website of DALL·E:
It’s amazing how DALL·E is able to create anthropomorphized versions of animals and objects and combine unrelated concepts in plausible ways!
For now, it’s not available to the public yet and is being tested will a small, controlled group of users selected from their waitlist. However, there are other open-source alternatives that are trained on a smaller amount of data to reimplement the original DALL·E neural network. One of them is DALL·E Mini, which still generates pretty amazing results despite being trained on a smaller dataset. This DALL·E Mini model is available for public usage by running the backend using Google Colab.
Removal.AI is an AI-powered tool that enables users to remove and change the background of an image by detecting the foreground pixel and separating the background completely from the foreground using advanced computer vision algorithms.
The background of a photo can be removed with just a few lines of code using the Removal.AI API. It can be integrated easily by just calling the API and passing the URL of the image as its payload and the image will be returned with its background removed. The API currently only supports background removal while background replacement can only be done through their website. However, they support custom needs if more functions are needed for the API integration. This API is very useful for image editing software companies to integrate background removal functions into their apps as it can be done with just an API request.
Bannerbear is an end-to-end solution for automated image and video generation for businesses of all sizes to automate and scale their marketing operations. You can auto-generate social media visuals, e-commerce banners, dynamic email images, and more through their API or no-code integration for popular softwares like Zapier, Airtable, and Make. Simply create a reusable template for the image using their template editor and send a request to their API endpoint to generate single or multiple images of different sizes:
Every text and image in the template becomes an object that can be easily modified depending on the parameter of the request sent. The text will also be auto-resized when you generate variations with long titles, product names, etc. To showcase how it works, here’s an image generation demo and API console demo on their website for you to try out.
Do not worry if you have faces on your images and think they might get cropped out. Bannerbear uses an AI-powered "smart crop" technology to auto-detect faces in photos and position them correctly in the output image. You can choose to have the faces automatically centered or offset them from the center, depending on the design of your template. Here's a comparison of an image generated using a regular crop vs. a smart crop:
The Bannerbear API can be used asynchronously or synchronously via different base URLs depending on your needs. In addition to the REST API, they also have official libraries in Ruby, Node, and PHP to make the integration into your existing platform or app easier. The API service starts from a free trial of 30 API credits and provides up to 50,000 API credits for a paid plan, which let you generate up to 50,000 images per month.
Bonus: The Cat API
Here is an image API that is different from the previous APIs but I think it is something fun and interesting to share. The Cat API is a “Cats as a Service”, as described by the company themselves, that provides a public service API all about cats. It is free to use when making your app, website, or service and has a paid subscription plan for $10 per month. The Cat API is developed by an API development agency that has created other API services like The Dog API. Despite being a rather niche API, The Cat API has delivered billions of requests to over 30K developers since its launch in 2012.
(Fun fact: other than images, the paid version of the API also has curated facts, jokes, and quotes about cats!😺)
Thanks for reading! I hope you find this article helpful. Feel free to share this article with your friends if you think they will enjoy reading this.
If you are interested in reading more image generation articles, here’s a tutorial on how to build a social media image generation app using Ruby on Rails.