AI photo editing software’s contains sets of instructions about the optimal quality of an image, in terms of content, reproduction, density, lighting and even cropping. These instructions have been assembled under a specific set of algorithms; the majority of these sets of instructions rely on 150 years of photographic experience and quality monitoring.
For example, the Luminar AI team, in their resource documents, indicates they collaborated with top photographers to train their software neural network, so it is almost like each image gets analyzed by this team of photo experts. However, because of the nature of photography and the aesthetics of imaging, the final result may not be pleasing to the user and the final decision to accept the correction needs to be made per image, as the software may misinterpret the image content and its desired impact in the storyline.
What about one-click image selection for a photobook, for example? Selecting photos and completing a personalized project instantly is the name of the game for retailers and consumers alike. For complex multi-image projects, the decision tree is not only about image selection, but also about coherence and aesthetics, an already daunting task for humans and even more for a machine. For example, at Mediaclip, we want to ensure the best experience for our users, and ensure that they have the highest degree of control, transparency, and efficiency while working on complicated high-engagement projects like photobooks. AI and Machine Learning should eventually help the creative process, but not take it over.
Why? Because inserting high-level and complex algorithms for best image selection, pleasing content, and accurate storytelling require special computer analysis, multiple scans, and classification of each image in a predetermined taxonomy. These efforts are lengthy; data analysis is time-consuming and requires costly computing resources at present.
A substantial set of diversified instructions need to be integrated (time-line, GPS, metadata, face recognition, deciphering between useful and useless images), all of which must be evaluated from a large set of the consumer image data to propose a pleasing and efficient “one-click” automated image selection. Currently, most eCommerce sites do not have these resources. Nick Burns, a Data Scientist, contends that no matter how great AI models are, they are merely at par with the data that is available to them.
Therefore, using AI and Machine Learning to perform image content selection, organization, and classification on a small and even large set of images that will convey the exact intended emotion and story can still be quite challenging.
For example, in a series of portraits, face recognition is important without a question. But is the subject winking or blinking? Is the subject well-postured? Or, is everyone smiling in a group photo? If not, then which image is the best one to select? In other circumstances, sets of criteria that provide the “knowledge” to the AI analysis may not reflect the perception of the actual moment/memory. An imperfect photo taken in low light conditions may carry the right feeling and should not be rejected or auto-fixed. Aesthetically, certain “imperfections” give character/individuality to an image and should not be softened. Because of nuances, proposed selections and image corrections still need to be confirmed by the user. Now imagine the challenge of selecting and organizing hundreds of images from multiple sources under one roof to construct one scenario.
As part of our research and development efforts, we are experimenting with AI abilities for sequencing, making scenarios, and other artifact integration possibilities of these sophisticated algorithms from an economic and technological perspective. Our process aims to achieve optimal one-click image selection and is regularly under review to ensure it offers the best possible scenario to our customers, their consumers, and the overall process.
Currently, our focus lies in improving the algorithms that simplify and optimize the content of complex page layouts. In our analysis, this yields a higher ROI for the business and a more intuitive end-user experience.
In summary, AI Deep Neural Networks for complex image projects, like photobooks would require years of data analysis about users’ perceptions of images or memories, as well as an intuitive AI index about their intent and behavior to be efficient in laying out complex one-click image and project suggestions.
The layers and profoundness of a human mind when making decisions (among other things) are unquestionably more sophisticated than AI and are unlikely to be replicated easily in the near future.
Like AI pioneer Geoffrey Hinton said: “I have always been convinced that the only way to get artificial intelligence to work is to do the computation in a way similar to the human brain. (…) We are making progress, though we still have lots to learn about how the brain actually works.”
