Was it a phone camera?
Not telling u,ll notice its the rear arch that went for a curve not the SkipperTomfoolery wrote:Looks like beer goggles to me.
Or scotch goggles?
Classic!!!With a GoPro, they don't read the sensor in top-down linear order (the way they do on a DSLR), they read it in a pattern that makes the next line read the farthest from the previous line read that still hasn't been read yet. This is the same pattern that describes how men select a urinal in a bathroom.
mastreb wrote:That artifact is caused by the read pattern of Go-Pros and other small-sensor video cameras. I see it all the time in my motorcycle videos, and it's quite annoying.
When you take a picture with a CMOS sensor, there's a limit to how fast you can read the image data out of the sensor chip itself that is based on the heat generated by the act of reading. Too much heat on a CMOS sensor increases noise, which shows up as sparkles and random color dots on photos. You've probably seen this in digital photos taken at night.
With a high-end DSLR, they can get incredibly high-resolution photos by reading the sensor somewhat slowly--over the course of 1/10th or 1/5th of a second, which is one of the limitations to shutter speed on a camera.
But with these same cameras when you switch to HD video mode, the resolution drops from 20 megapixels to HD, which is just 2 megapixels, because they now have to read that sensor 30 times per second. So they literally only read every 10th line or so in order to spread out the system heat.
With a GoPro, they don't read the sensor in top-down linear order (the way they do on a DSLR), they read it in a pattern that makes the next line read the farthest from the previous line read that still hasn't been read yet. This is the same pattern that describes how men select a urinal in a bathroom.
When horizontal objects are moving across when a picture is taken, that read pattern causes the artifact you see here, because the image shows up at slightly different times based on the read moment of that line.
It's either camera movement or subject movement. They're equivalent for this problem.seahouse wrote: So it's camera movement then?
Correct.In this case, the alternative to a wavy picture would be a picture with motion blurr instead, because the image stabilization has minimized that (the image is fairly sharp, in addition to the higher shutter speed) and at the same time revealed the existence of the waviness. Cake and eat it... not?
Also correct. I suspect that they alternate left-right/right-left reading to distribute read heat as well, although I don't know that.There is also some distortion in the vertical direction as well as the horizontal, because there is also a read-time lag for each scan in the left-right direction. Otherwise with the field scanned in wide horizontal stripes, all the waves would align in all the vertical elements (mast, stanchion, jib), which they don't in the picture.
Such a simple solution! They orient the reading this way to prevent horizon waviness, which is a factor in nearly every outdoor picture and much more common than vertical edges.So I would suggest a solution in this case would be to turn the camera 90 degrees, then take the picture, whenever the picture is dominated by vertical elements (such as this picture is). And leave the camera in the normal orientation when the picture is dominated by horizontal elements, if you wanted to reduce waviness in these conditions.
- B.
That, and with sensor chips that have an aspect ratio that is not square, fewer scan lines are required to cover the shorter (vertical) length than the longer (horizontal) length.They orient the reading this way to prevent horizon waviness, which is a factor in nearly every outdoor picture and much more common than vertical edges.
, and not taken in 2013 either 
. Sorry. 
