16 ms·

Does the photograph show the system that is described? The system has 'forward-facing cameras mounted on the side mirrors and front grille'. But the second ph

Does the photograph show the system that is described? The system has 'forward-facing cameras mounted on the side mirrors and front grille'. But the second photograph seems to show what is underneath the car, which is not a view that's possible from those cameras at that moment in time. There's no shadow, either. Is it possible to achieve this just by delaying the image feed? edited

RodgerTheGreat 7y ago
I'd imagine that if it works as shown in the promo photos it's extrapolating from a time-lagged version of the front view. Seems potentially misleading to users; As a driver I'd rather not be shown extra information if it's a faked guess.
- ASalazarMX 7y ago
  It clearly warns "Drive forward to view underneath bonnet" and shows a black area before the stitching. As soon as it starts stitching, it warns "View underneath bonnet not live". This is clear enough to not be confused with real-time.
idiot900 7y ago
From the article: "Computer controls slightly delay the images from the front camera and stitch them together with side-mirror views to create a seamless real-time feed." Seems like it just does an affine transformation on the image of the ground ahead as the car moves forward to fill in the missing parts of the image, with a time delay scaled inversely with the speed of the car. I wonder how (or if) it deals with non-static objects the car is driving over, such as moving water.
saulrh 7y ago
Not really possible to do with a simple delay, and if it was that easy it'd have already been done. However, the math you do need is a close relative of the panorama mode you have on your phone, and I feel like it's about the right time for it to have been refined and hardened well enough to show up on a car. Conceptually it's simple: Take the images, stitch them together, center the view on the cameras. Repeat with the next frame, but instead of using only the new images, stitch them into the same panorama you started building last frame. Keep stitching even as older images start to move entirely out of the cameras' field of view. Garbage-collect eventually, but you have a constant target synthetic FoV so you can guarantee bounded consumption. Add a few years of engineering and you have a really limited, single-propose SLAM system cobbled together from OTC algorithms that's just good enough to handle the ground underneath an SUV.
netsharc 7y ago
https://youtube.com/watch?v=o7V5COxB48Q https://youtube.com/watch?v=o7V5COxB48Q here's a demo, from 4:53 onwards. Yeah it's delayed, so if e.g. a cat moves after it is out of sight of the camera, the picture won't display the actual position of said cat.
- FatalLogic 7y ago
  That's useful, but like other people said, it could be misleading. If the front wheel sinks into soft ground when you're off road, the display can't show you that, then you could high center the vehicle and get stuck Or you could drive over your cat
  - burfog 7y ago
    I wonder if a pair of cameras under the center of the vehicle would meet the requirement for a rear-view camera. That'd be superior, because seeing right next to the wheels is actually useful. For viewing higher, we have windows and mirrors.
- joking 7y ago
  many cars have the same to show the sides of the vehicle using only the front and rear cameras without the need to have cameras also in the side mirrors. It creates a texture with the cameras and use it to render the surroundings when you move around. Pretty cool when I saw for first time. But yes, if something moves where the cameras are not pointing you are not going to see, my vehicle only moves forward and rear, so it's pretty useful anyway

RodgerTheGreat 7y ago

I'd imagine that if it works as shown in the promo photos it's extrapolating from a time-lagged version of the front view. Seems potentially misleading to users; As a driver I'd rather not be shown extra information if it's a faked guess.

ASalazarMX 7y ago
It clearly warns "Drive forward to view underneath bonnet" and shows a black area before the stitching. As soon as it starts stitching, it warns "View underneath bonnet not live". This is clear enough to not be confused with real-time.

idiot900 7y ago

From the article: "Computer controls slightly delay the images from the front camera and stitch them together with side-mirror views to create a seamless real-time feed." Seems like it just does an affine transformation on the image of the ground ahead as the car moves forward to fill in the missing parts of the image, with a time delay scaled inversely with the speed of the car. I wonder how (or if) it deals with non-static objects the car is driving over, such as moving water.

saulrh 7y ago

Not really possible to do with a simple delay, and if it was that easy it'd have already been done. However, the math you do need is a close relative of the panorama mode you have on your phone, and I feel like it's about the right time for it to have been refined and hardened well enough to show up on a car. Conceptually it's simple: Take the images, stitch them together, center the view on the cameras. Repeat with the next frame, but instead of using only the new images, stitch them into the same panorama you started building last frame. Keep stitching even as older images start to move entirely out of the cameras' field of view. Garbage-collect eventually, but you have a constant target synthetic FoV so you can guarantee bounded consumption. Add a few years of engineering and you have a really limited, single-propose SLAM system cobbled together from OTC algorithms that's just good enough to handle the ground underneath an SUV.

netsharc 7y ago

https://youtube.com/watch?v=o7V5COxB48Q https://youtube.com/watch?v=o7V5COxB48Q here's a demo, from 4:53 onwards. Yeah it's delayed, so if e.g. a cat moves after it is out of sight of the camera, the picture won't display the actual position of said cat.

FatalLogic 7y ago
That's useful, but like other people said, it could be misleading. If the front wheel sinks into soft ground when you're off road, the display can't show you that, then you could high center the vehicle and get stuck Or you could drive over your cat
- burfog 7y ago
  I wonder if a pair of cameras under the center of the vehicle would meet the requirement for a rear-view camera. That'd be superior, because seeing right next to the wheels is actually useful. For viewing higher, we have windows and mirrors.
joking 7y ago
many cars have the same to show the sides of the vehicle using only the front and rear cameras without the need to have cameras also in the side mirrors. It creates a texture with the cameras and use it to render the surroundings when you move around. Pretty cool when I saw for first time. But yes, if something moves where the cameras are not pointing you are not going to see, my vehicle only moves forward and rear, so it's pretty useful anyway