The past few years will probably be remembered as those in which artificial intelligence (AI) became accessible to the general public. Numerous AI tools that can make life easier are currently available to ordinary people. Some of these mind-blowing AI tools are presently free, or free to use, for a limited time or for a limited amount of work.
Image creation and manipulation
Many people, especially those who have lived through a few changes in photographic technology, have some old photographs that have discoloured over time. This is where the Generative Facial Prior-Generative Adversarial Network (GFP-GAN) free AI tool can help – it can restore these photos to their former glory.
GFP-GAN was developed by the Chinese company Tencent, in which Naspers is a significant shareholder. It merges two AI models to replace the missing detail in a photo in seconds. The result is a restored photo with a good balance between realness and authenticity. GFP-GAN blind face restoration restores facial details and enhances colours with just a single forward pass. The AI model works with even very low quality and low-resolution files, and easily outperforms many photo restoration AI tools to deliver incredible results.
Due to copyright limitations on the use of several images available on the Web, it has become important to create your own images. DAL. E 2, developed by Open AI (the developers of the famous ChatGPT), is a free AI system that creates photo realistic images and art based on a description in any natural language.
The high-resolution images and art that are created from a text description are very realistic through the combination of attributes, concepts and style. DAL. E 2 can also retouch photos, replace objects within the image with AI-generated imagery or inpainting, and blend AI-generated images into real photos or augment them with AI-generated content. When creating an image, it can also provide variations based on different angles and styles. DAL. E. 2 certainly helps people to express themselves visually.
When we look at old family photos, we sometimes wonder how these ancestors moved, smiled and looked in real life. It is now possible to see this, with Deep Nostalgia Photo Animation by MyHeritage, a company specialising in video re-enactment though the use of deep learning.
This technology makes it possible to animate the faces in historical photos and create high-quality, realistic video footage of how a person from an old photo could have moved and looked if they were captured on video. The Deep Nostalgia™ feature uses several drivers. Each driver is a video consisting of a fixed sequence of movements and gestures. Deep Nostalgia™ can very accurately apply the drivers to a face in a still photo, creating a short video that can be shared with friends and family. The driver guides the movements in the animation and enables the ancestors to smile, blink and turn their heads, thus bringing the photos to life!
Video creation and editing
Video has become the most popular medium of our time and therefore became an important communication and marketing tool. However, creating a good video requires knowledge, skills and the right equipment. Well, that was until recently, before powerful AI video editors and generators were introduced.
Often marketers and publishers wish that video creation was easier or that there was a way that articles, blog posts or meetings could be transformed into social videos.
Lumen5 video creator is an online video-creating platform with a variety of tools that makes this possible. It enables the user to create engaging video content with ease. It includes several templates and hundreds of customisable designs that can be used to communicate effectively and consistently depending on the specific social media platform. The library contains numerous images, video footage and music that can be used in social posts. The user only needs to import a transcript upon which the AI tool will generate a whole video sequence based on the images in the transcript.
A whole range of images and video clips that can be included in the final video is offered. What is special about Lumen5 is that it makes the creation of a video easier than building a presentation. It automatically turns blog posts into video, or transforms zoom recordings into captivating video clips. It cuts the traditional agency time of about one week for the production of a video down to a few hours.
Synthesia.io, an amazing video generator, allows the creation of videos with human presenters in minutes by simply typing in plain text. There are more than 100 diverse AI avatars based on real actors that can speak text in 120 languages. Synthesia makes it possible to easily turn boring documents, PowerPoints and PDFs into engaging training, how-to or product marketing videos. It is also possible to use Synthesia to turn text into professional voiceovers without using a microphone.
Since Synthesia AI voices are digital clones of the voices of real people, the recording of a voice is no longer needed. One training company found that by using Synthesia to develop a virtual facilitator to guide learning through a training session, it increased the engagement of the learners by 30%.
Video editing is known to be a tedious and lengthy process and usually requires a significant level of technical skills and expensive software. AI tools dramatically changed these requirements. Runway, a next-generation content creation suite, using generative AI, makes editing extremely easy. It has the ability to enhance existing video footage by removing backgrounds, painting out objects, motion tracking, text to image generation, image-to-image generation, erase and replace, cleaning audio, animation and the automatic adding of subtitles. It offers more than 30 AI-powered Magic Tools to help with ideation and generation, real-time video editing, and collaboration. And everything runs inside the browser and does not require any downloads.
Due to the popularity of video, one problem that we all experience is that the amount of videos on our computers have increased dramatically over the past few years. This makes the finding of the right video often extremely difficult. Luckily, the very useful Nova AI was introduced.
Nova AI is a web-based video categorisation tool that assists users to find and categorise video more easily. Nova AI uses AI to recognise, log and tag different elements within the videos such as people, faces, expressions, emotions, objects, activities, celebrities and many more. This makes it a great tool for organising all kinds of video collections. But Nova AI is much more. It is also a full-scale video editor that automatically transcribes, translates and adds subtitles to video content.
The AI recognises human speech patterns to generate subtitles and transcriptions that can be auto-translated into more than 40 languages. Nova AI can further auto-generate shorter or bite-sized video clips for digital projects and media campaigns based on longer videos in the content library. It is even possible to edit YouTube and TikTok videos just by grabbing the link without downloading the video.
The power of AI
AI is now everywhere and can make our lives much easier, especially in the fields of image and video generation and editing. Not only does AI software enhance our productivity and significantly save costs by enabling in-house production, but it empowers people and organisations to express themselves creatively.
But is also clear that the lines between reality and creativity are becoming increasingly blurred.
Professor Louis C H Fourie is an Extraordinary Professor in Information Systems of the University of the Western Cape.
BUSINESS REPORT