We’ve come a long way since cameras were first integrated into cell phones around a quarter of a century ago. Primitive by today’s standards, those low-resolution cameras lacked even basic features like autofocus and flash and produced grainy, low-res images that looked like a blurry mess on any screen bigger than your phone’s.
By Rami Osman, director for business development at MediaTek Middle East and Africa
Today, by contrast, the cameras on mid-range to high-end smartphones offer nearly anyone the ability to capture pro-level images and video. One of the key technologies that has transformed the videography and photography experience over the past few years is the integration of artificial intelligence (AI) into your smartphone.
From enhancing low-light exposures to reducing blur in action shots, AI enables even amateur shutterbugs to take high-quality photos with minimal effort. Much of the magic is to be found in a system-on-chip (SoC), like the MediaTek Dimensity 7300 or our recently launched MediaTek Dimensity 9400.
This important piece of silicon integrates the central processing unit (CPU), graphics processing unit (GPU), and Neural processing unit (NPU) that your phone uses to run complex tasks. The SoC brings together dedicated AI, imaging processors and accelerators that help you take amazing photos and images.
Here are some AI camera features that many smartphone users already take for granted:
- Smart HDR – AI gives a significant boost to High Dynamic Range (HDR) photography. Our HDR-ISP Image Signal Processors work with the in-chip AI processors to enhance HDR capture with a range of post-processing optimisations. This enables smartphones to produce photos with greater detail and clarity in all lighting conditions. Our AI-Camera also provides features such as image construction, down to pixel-level enhancements.
- AI noise reduction – Low-light photography is challenging at the best of times. AI cameras in smartphones can brighten images, reduce noise, and enhance details even in poor lighting conditions. AI noise reduction techniques accurately mitigate ISO noise to provide consistently better results that retain fine details, particularly in low-light environments.
- Low-light videography – Users can now readily capture 4K HDR video in low-light situations thanks to AI-NR that’s now faster and more accurate than ever before, without the smartphone slowing on extended use or eating into battery life. AI-auto white balance with high contrast compensation, AI-auto exposure, zero lag AI-shutter, and AI auto-focus enable clear video capture with high detail and no blur.
- Motion unblur – NPU-ISP technologies can recognise the movement of a subject in a shot and automatically determine the best settings in real-time. The chip will capture multiple simultaneous frames at different exposures, ensuring any fast action is a sharp and clear snapshot of the exact moment.
- Scene and object detection – AI-powered smartphone cameras can detect different scenes automatically, such as landscape, portrait, night shot, or macro image, as well as objects such as people or pets. They can then optimise exposure, contrast, and saturation to ensure the best shot. AI can also distinguish between the subject and the background to apply a natural bokeh effect.
- Gen-AI Super Zoom – The MediaTek Dimensity 9400 with Imagiq 1090 provides powerful tools to creators. It enables HDR video capture throughout the entire zoom range, mimicking a professional camera zoom lens, with a negligible drop in quality. In addition, the Dimensity 9400 SoC can use Generative-AI Super Resolution to click sharp photos of far-off objects. The feature allows users to achieve 100x magnification by filling the blocky or pixelated bits with corrective fill that can produce resoundingly vivid results.
The future: Advanced computational photography and AR
We are only just starting to see the power of AI photography come to life. In future, advanced computational photography will take features like multi-frame processing, noise reduction, and super-resolution to a whole new level. This will allow sharper, more detailed images.
AI cameras will better recognise scenes, objects, and specific subjects, adjusting camera settings automatically for optimal results. Future AI cameras will continue to improve video capture, providing tools like automatic subject tracking, intelligent zoom, and stabilisation.
Generative AI will soon start to come into its own. As prefaced previously, super zoom with Gen AI lets you smooth over the rough edges of pixelated photos. Generative AI may also suggest optimal framing and composition techniques based on rules of photography, helping you take more aesthetically pleasing shots.
The AI camera will also be central to next-generation augmented reality (AR) and virtual reality (VR). AI enables these technologies to overlay digital content seamlessly onto the real world, which is particularly useful in immersive applications such as virtual clothing try-ons, interactive games, and educational tools.
Whether you’re a professional photographer, media influencer or just a regular user recording your memories to share with friends and family, your smartphone’s AI camera will help you take better, more creative pics and videos. At last, nearly anyone can enjoy near-DSLR results from a camera they carry with them everywhere they go.