MobiusStrip
New member
Hi all.
I read a few articles about this issue and looked at a bunch of footage. It seems to me that there are some suspect "explanations" floating around.
The HD-sized images from these cameras are being downsized from a vastly larger frame; I'd think any aliasing due to the chip and its OLPFs would occur at the native resolution of the chip. Resizing it down to HD resolution would reduce (if not totally obliterate) this extremely fine aliasing. Barry notes this effect too:
“...since web video is frequently scaled down in size (and scaling down helps to reduce aliasing artifacts) they released the 5D Mk II as-is...”
Scaling an image larger than 5000 x 3000 down to 1920 x 1080 should not result in more aliasing than a full-sized still. And 1280 x 720 should look smoother still. But in fact we don’t see that at all; most people think the 720 footage looks MORE aliased (although that’s probably because they’re blowing it up to the same size as the 1080 for viewing).
The aliasing in DSLR video almost certainly comes from one thing: a crappy resizing algorithm. What happens if you resize an image with no resampling? Try it in your favorite image-editing program. The result: wretched aliasing.
To achieve the necessary processing throughput to deliver “HD” video without cropping the sensor, these cameras are taking shortcuts. They’re not downsampling the images well, and the result is aliasing galore. They may even be doing some abbreviated deBayering to speed things up. I question the idea that somehow the aliasing occurs because the chip is "not optimized for video." The limitation here is computing power and data throughput. As we all know, these are likely to improve over time. My guess is that the images from the exact same sensor, run through a good-quality downsizing algorithm, would match anyone's expectations for HD. I also think that even with a more aggressive OLPF array on the sensor you'd still see aliasing, thanks to the bad downsizing.
I read a few articles about this issue and looked at a bunch of footage. It seems to me that there are some suspect "explanations" floating around.
The HD-sized images from these cameras are being downsized from a vastly larger frame; I'd think any aliasing due to the chip and its OLPFs would occur at the native resolution of the chip. Resizing it down to HD resolution would reduce (if not totally obliterate) this extremely fine aliasing. Barry notes this effect too:
“...since web video is frequently scaled down in size (and scaling down helps to reduce aliasing artifacts) they released the 5D Mk II as-is...”
Scaling an image larger than 5000 x 3000 down to 1920 x 1080 should not result in more aliasing than a full-sized still. And 1280 x 720 should look smoother still. But in fact we don’t see that at all; most people think the 720 footage looks MORE aliased (although that’s probably because they’re blowing it up to the same size as the 1080 for viewing).
The aliasing in DSLR video almost certainly comes from one thing: a crappy resizing algorithm. What happens if you resize an image with no resampling? Try it in your favorite image-editing program. The result: wretched aliasing.
To achieve the necessary processing throughput to deliver “HD” video without cropping the sensor, these cameras are taking shortcuts. They’re not downsampling the images well, and the result is aliasing galore. They may even be doing some abbreviated deBayering to speed things up. I question the idea that somehow the aliasing occurs because the chip is "not optimized for video." The limitation here is computing power and data throughput. As we all know, these are likely to improve over time. My guess is that the images from the exact same sensor, run through a good-quality downsizing algorithm, would match anyone's expectations for HD. I also think that even with a more aggressive OLPF array on the sensor you'd still see aliasing, thanks to the bad downsizing.
