Q: Please briefly introduce yourself, Samsung R&D Institute Japan-Yokohama, and the kind of work that goes on there. What projects are you working on?

Samsung R&D Institute Japan-Yokohama (SRJ-Y) was established in 1997. We contribute to the Device eXperience (DX) division’s Digital Appliances (DA), Visual Display (VD), and Mobile eXperience (MX) businesses, as well as Samsung Research in fields such as materials, vital core parts, manufacturing processes, and software development, leveraging Japan’s strengths in these areas. With Japan having a concentration of world-class camera manufacturers, high-quality camera technology is also one of our strong points.

I am Shinri Takahashi, leading the development of high image quality for mobile phone cameras at SRJ-Y. Joined SRJ-Y in 2010, I currently lead a project team that develops new algorithms to further improve Auto Exposure (AE) and Auto White Balance (AWB) technology after the development of digital cameras.

Q: Please tell me about the importance of your research field or technology.

Mobile phones have many features, but the camera is one of the most critical aspects that make one stand out from competitors. The ability to capture excellent image quality is directly linked to the distinctiveness of mobile phones. Camera image quality involves various factors, including brightness, color, contrast, resolution, noise, dynamic range, etc. Among these, brightness and color, which significantly influence human senses, form the basis of AE/AWB technology. This is an essential technique because even if all other factors are at the highest level, the image quality will deteriorate if the brightness and colors are unsuitable. And this is also a complex technique, for instance, in the case of noise, the smaller the size, the better the image quality, except in special cases. However, there is no definitive answer regarding brightness and color. Target values are suitable for various situations and vary according to individuals’ preferences. As such, maintaining consistent control over not only the target value but also the shift in the control value when the scene changes is crucial to avoiding user dissatisfaction.

These two technologies directly affect human emotions. Therefore, they have not been made publicly available through open source in recent years and remain within each company’s proprietary knowledge.

Q: Can you tell us about any main achievements, rewarding moments, or any significant events in your research?

Ever since I was a child, I have loved railways and cameras, and my hobby was taking pictures while traveling using my father’s camera. Back then, I used a film camera, but when I developed and looked at the photos, it was challenging to get satisfactory results. Fortunately, now I am responsible for developing mobile phone cameras after working with film and digital cameras. After implementing the developed algorithm in the development model, we conduct a field test. I feel satisfied when I achieve better image quality compared to previous algorithms or competitors’ products, and even if there is an aspect that requires improvement, being able to enhance it on my own is truly rewarding. Seeing my developed algorithms being commercialized and released to the world brings me immense pleasure.

One of the difficulties of the Digital Camera AE control has been its issue with the Image Sensor’s limited dynamic range. If you try fully express the dark parts in a scene with a substantial difference in terms of brightness, the bright areas will turn white. In contrast, if you adjust it to avoid having white areas, the dark part will become darker. Even if this is due to technical constraints, I want to resolve it from the users’ perspective.

Over the past few years, there have been significant improvements in high dynamic range (HDR) processes that synthesize multiple images with varying exposures. However, this method cannot be applied to the preview image before capturing. Nevertheless, preview images have also experienced considerable enhancements by adopting HDR effects using a different approach than static images. It is satisfying to witness a problem that I have been working on for many years finally resolved.

Q: What is your vision for the future, and what goal would you want to achieve?

As mentioned earlier, there is no definitive answer regarding the brightness and color of a photo. Until now, I believe that the goal of this technology is to satisfy as many people as possible. However, the absence of a clear-cut solution suggests limitless possibilities. Even now, there are ongoing research studies into catering to individual preferences, which are subject to change over time.

With the advancement of hardware and artificial intelligence technology in recent years, possibilities deemed impossible until a few years ago are now within reach. As such, my goal is to spearhead groundbreaking developments in high-quality imaging that transcends previous limitations.

Q: What advice would you give to aspiring researchers interested in pursuing a career in camera image quality?

A comprehensive knowledge of not only software technology, but also physical and hardware knowledge, such as what light is, how sensors turn it into signals, and how images are processed, would be a great help.

Many people use cameras to record happy moments. The awareness of how to support them in preserving those memories as beautifully as possible will help maintain your motivation at a high level.