AI has also been transforming various fields and industries, including photography.
Artificial intelligence (AI) is a branch of computer science that aims to create machines and systems that can perform tasks that normally require human intelligence, such as speech recognition, image recognition, and decision making. AI has been advancing rapidly in recent years.
Photography has evolved significantly since its invention in the 19th century, with the development of new technologies and techniques, such as film, digital, and mobile photography. However, the advent of AI has brought a new level of innovation and possibility to photography, both in terms of image capture and image editing.
In this blog post, we will explore some of the ways that AI is changing photography and what it means for photographers and consumers.
AI in Image Capture
AI can enhance the process of capturing images, by analyzing and optimizing various aspects of a photograph, such as scene recognition, subject tracking, and exposure control. AI can also enable new features and functions that were not possible before, such as night mode, bokeh effect, and AI-generated images (the new Google Pixel phones are an example of this with their new Face-Swap feature).
One of the applications of AI in image capture is scene recognition, which is the ability of a camera or a device to identify and classify the objects and elements in a scene, such as people, animals, plants, buildings, etc. Scene recognition can help a camera or a device to adjust the settings and parameters automatically, such as exposure, white balance, focus, and colour to achieve the best results. Scene recognition can also help a camera or a device to provide suggestions and feedback to the user, such as composition, framing, and lighting, to improve the quality and aesthetics of the image. For example, a camera or a device can detect if the horizon is level, if the subject is in focus, or if the lighting is sufficient, and prompt the user to correct or enhance the image.
Another application of AI in image capture is subject tracking, which is the ability of a camera or a device to follow and focus on a moving subject, such as a person, an animal, or a vehicle in a scene. Subject tracking can help a camera or a device to capture sharp and clear images of a subject, even in challenging situations, such as low light, fast motion, or complex backgrounds. For example, a camera or a device can lock on a subject and maintain the focus and exposure, even if the subject moves across the frame or changes the distance from the camera or the device. A good example of this is the AI Tracking found in OM System camera bodies that can recognise certain subjects and focus on the correct part of that subject automatically, for example when it recognises a car it tries to find and focus on the driver, or with an animal it will look for eyes.
Night mode is a feature that allows a camera or a device to capture images in low-light or dark conditions, without using a flash or a tripod. Night mode uses AI to combine multiple images with different exposures and settings to create a single composite image with improved brightness, details, and colours. Night mode can also use AI to reduce noise, blur, and artefacts that are common in low-light or dark images, and to enhance the sharpness, contrast, and saturation of the image.
Bokeh / Lens Blur Effect
Another example of AI in image capture is the background blur effect, also known as lens blur or bokeh. This can create a pleasing and artistic effect, by isolating the subject from the surroundings and by highlighting the shape and colour of the subject. Portrait Mode on most mobile phones is a perfect example of this, where it can create a professional or cinematic effect by simulating the look and feel of a DSLR or a mirrorless camera.
This works by using AI to detect and segment the subject and the background in an image, and to apply a variable blur to the background and foreground, depending on the distance and the shape of each element. You can often adjust the intensity and the quality of the blur to achieve a particular look.
And now for the most contentious aspect of AI in photography, AI-generated images that are created by AI, without using a camera or a device. Using AI to create new images from scratch or to modify existing images based on a text prompt, a sketch, or a style, AI-generated images can be realistic, surreal or artistic.
AI-generated images use AI models and algorithms to learn from large datasets of images to generate new images that are similar but not identical to the original images. AI can also combine and manipulate different elements and concepts in an image, based on text prompts, enabling users to create images that are not limited by the physical world.
This presents a dilemma for some photographers as it is easy to blur the lines between what is real and what is imagination.