Converting a Video to 4K: How It Works Explained

The transition from analogue film to digital video technology occurred rapidly. In a short span, traditional tape-based video recorders were replaced by digital video cameras, and eventually, smartphones took over. Today, 4K resolution has emerged as the gold standard for video production, while 1080p serves as its lower-quality counterpart.

Despite the widespread adoption of 4K, many videos are still produced in 1080p. a vast amount of older content, including home videos and classic films, exists in standard definition or even lower resolutions. This reality prompts many to consider converting their videos to 4K, aiming to enhance their appearance on modern 4K OLED displays. Although converting video to 4K is feasible, it involves a complex process with several critical considerations.

When discussing 4K and 1080p, or ultra-high definition versus high definition, we are referring to the number of pixels contained within each frame of video. For ultra-high definition content, commonly known as 4K, the resolution is 3,840 x 2,160 pixels, compared to 1,920 x 1,080 pixels for 1080p high definition video. In contrast, standard definition video typically has either 480 or 576 vertical lines, with the width varying based on the original filming format. cinematic 4K, often referred to as ‘true 4K’ or DCI 4K, features a resolution of 4,096 x 2,160 pixels.

Upscaling any output file to 4K is possible since 4K simply refers to frame size. To convert from 1080p to 4K, you need to add three additional pixels for every original pixel recorded, resulting in a total of four times the number of pixels. achieving this without sacrificing quality is a significant challenge. Merely changing the resolution does not uncover hidden clarity or sharpness.

Traditional upscaling software relies on interpolation to manage the scaling process. It increases each frame’s size and attempts to determine the appearance of new pixels. The effectiveness of the upscaling software is a crucial factor. It estimates how the new pixels should look, but it cannot retrieve genuine detail that was never captured in the original footage.

This often results in upscaled footage appearing softer. The estimation process used to generate additional pixels can create a smoother appearance. While sharpening can enhance edges, it presents a challenge: finding the right balance is essential, as too much or too little can compromise the final output. Noise reduction poses a similar challenge, potentially resulting in an unnatural appearance in the final video.

For professionals aiming to convert video to 4K, accessing the highest quality source data is crucial. The quality of the source files directly impacts the final output quality.

The HD restoration of Star Trek: The Next Generation exemplifies the challenges involved in achieving a successful upgrade. Commissioned by CBS in 2011, this restoration utilized the show’s original 35mm film negatives, which provided a quality comparable to a 20-megapixel resolution. The remastering process transformed all seven seasons from their original broadcast quality to 1080p for Blu-ray release, involving 25,000 reels of original film stock and enhancements in visual effects and CGI. Completing this meticulous project took over three years and reportedly cost CBS more than $12 million, according to one of its producers.

Ironically, the shift from analog film to digital video simplifies such restorations. Working with an older digital master file with a fixed resolution limits your ability to enhance the content. The original negatives used in the Star Trek restoration captured far more detail than what was visible in the original television broadcasts.

When dealing with source files of low resolution, the challenge becomes significantly more complex. Any additional pixels generated are merely educated guesses. The best possible outcome may look reasonably convincing, but it will not be entirely true to the original.

Unlike large studios with substantial budgets like CBS, home users typically do not have the luxury of $12 million to invest in converting their videos to 4K. Instead, they must choose between traditional interpolation techniques and innovative AI-based tools. Both methods generate the additional pixels necessary to fill out an enlarged 4K frame, but they operate differently.

4K upscalers using interpolation rely on established mathematical principles. They examine a pixel from a single frame and evaluate the color and brightness of adjacent pixels. Based on this analysis, they attempt to predict the appearance of surrounding pixels. While basic interpolation may involve simple duplication, more sophisticated algorithms aim to preserve edges and minimize jagged lines. The results can be somewhat predictable, but a common issue is that the footage may appear softer. Interpolation cannot recreate authentic details or textures, leading to a somewhat artificial look. Many modern TVs also include built-in upscaling capabilities, but their effectiveness varies.

AI upscaling offers a promising solution to these limitations. By training models on examples of both low and high-resolution images, AI can identify how common features should appear on your screen. Rather than estimating each pixel, AI predicts the overall appearance of a sharper image.

The outcome can be sharper and more realistic, but it still may not perfectly replicate reality. AI upscalers can introduce unique artifacts that traditional interpolation does not. Continuous improvements in AI models mean that the quality and consistency of results will likely evolve significantly in the coming years.

When working with older videos, consider interlacing. Older video frames were split into two halves by odd and even horizontal lines, refreshing each line with every screen refresh. Modern displays do not use interlacing, but older broadcasts meant for CRT screens did. To convert such videos effectively, you will need to de-interlace before upscaling. achieving satisfactory watchable quality when upscaling a 480i NTSC video to 4K can be challenging.

Upscaling tools typically offer a variety of settings to help you achieve your desired results, whether you choose AI or interpolation. If the initial attempt does not meet your expectations, you can adjust the settings and try again, provided you have sufficient disk space.

Simply inserting a 1080p video into an upscaler does not guarantee a flawless 4K video outcome. Technically, you can increase the video frame size, but achieving a smooth transition without sacrificing quality is far more complex. It may require multiple attempts to align the results with your expectations.

Success is more likely with high-quality source material, and smaller resolution jumps yield better results. For instance, upscaling from 1440p to 4K requires fewer ‘fake’ pixels compared to upscaling from 1080p to 4K or even lower resolutions like 480p to 4K. 4K upscaling software is fundamentally guessing the appearance of the missing pixels. While AI advancements are enhancing the process, a perfect match remains elusive.

Resolution is not the only key factor. To conserve storage space, video files may be compressed with lower bitrates, reducing the data available for every second of video. Upscaling to 4K cannot rectify this issue, and common compression artifacts such as blocks or banding cannot simply be discarded if you are working with a low-bitrate file. the quality of the source material is paramount; with a solid foundation, upscaling becomes significantly more manageable.