Oppo’s plan on perfecting photos and videos with AI: just embrace HVS (that’s not VHS!)

0comments
Oppo’s plan on perfecting photos and videos with AI: just embrace HVS (that’s not VHS!)
As of December 31, 2023, OPPO has filed over 3,160 global patents related to AI. That might sound like a lot, and 3,160 is a pretty hefty number – but if you want to stay in the AI craze race, you better be prepared to work hard.

Oppo, unlike Apple, is rejecting the idea of lagging behind Samsung and Google in the AI race, and has a thing or two to show off.

The brand that gave to the world the Oppo Find N3 (a.k.a. the OnePlus Open, a.k.a. a favorite of ours and probably the best foldable phone of 2023) is present at MWC and while they sure do present new devices, Oppo also shared some AI-related tools and platforms to further enhance the hardware they sell.

Oppo’s AI tools and platforms that we’re about to present are set to deal with something very annoying – blurred photos and videos.

As Oppo says:

.

Enter the AI Motion technology.

What’s EVS? What’s HVS? (Also, if you’re a juvenile: what’s VHS?)


Right off the bat: VHS was the closest thing to magic that 80s and 90s kids around the world had in their living rooms – a black plastic furnace, roughly the size of a laptop (but way thicker), that played movies (they also came in the form of a black plastic brick). Ah, those pre-Netflix dark-ages of home entertainment!

My rant is not totally off the subject, as anyone who had the honor of dealing with a VHS knows how questionable the picture quality could get. VHS tapes were prone to wear and tear, and much more: tracking errors, snowy picture patterns, color bleeding, ghosting, jitter, wrinkles, and more.

Recommended Stories
Of course, mobile technology has come a long way, and smartphone cameras can’t be accused of offering the same VHS picture problems in 2024, but they’re far from perfect.

In order to get them a step closer to perfection, Oppo in collaboration with AlpsenTek and Qualcomm, unveiled AI Motion technology at MWC to advance smartphone imaging capabilities, addressing the challenge of capturing clear images of fast-moving objects.

AI Motion employs Event-based Vision Sensing (EVS) and Hybrid Vision Sensing (HVS) technologies, enabling high-resolution, motion-blur-free photography and videography.

  • Conventional image sensors
Conventional image sensors collapse all motion information within a specific time period (i.e. the exposure period) into a single frame. This leads to motion blurs in images and loss of valuable motion data.

  • EVS (Event-based Vision Sensing)
EVS (Event-based Vision Sensing) is an imaging technology that continuously records motion information using a shutter-free mechanism. By responding only to changes in light intensity, EVS sensors can capture motion information with a high time resolution at a lower computational cost, allowing them to achieve a better balance between high frame rates and high dynamic range with low data redundancy and low power consumption. Despite its many benefits, EVS technology often lacks critical static visual information that is needed for many machine vision applications.
 
  • Hybrid Vision Sensing (HVS) technology
HVS combines EVS and conventional imaging technology to simultaneously output and process high-quality RGB images and EVS data, so that we can efficiently capture images with embedded motion information that is highly compatible with data processing. To better leverage the EVS data information for enhanced image quality, Oppo has incorporated AlpsenTek's hybrid vision sensor, ALPIX-Eiger, into smartphones and played a leading role in developing a set of AI Motion algorithms that can deblur images and support video interpolation for high-frame videography, enabling high-quality, real-time capture of high-speed moving objects. Thanks to the support from the Snapdragon Mobile Platform and motion data captured by EVS, the AI Motion algorithm can restore more details lost to motion in images with enhanced accuracy. With no need to sacrifice exposure time either, the solution delivers overall improvements in image quality through greater clarity and realism, regardless of whether it is the image subject or the phone itself that is moving.

The two sample photos below demonstrate the effect of de-blurring on image quality. The AI Deblur feature enables even rapidly moving subjects to be captured without ghosting or blurring, resulting in clearer images with more accurate detail.



Also, there’s support for high frame rate video interpolation, taking slow-motion videography to the new level. Unlike conventional constant time interval interpolation, the Slow Motion feature is realized by capturing motion information that can be used to recover movement trajectories more accurately, resulting in 960fps or higher slow motion video at a theoretical 2K/4K resolution.

Recommended Stories

Loading Comments...
FCC OKs Cingular\'s purchase of AT&T Wireless