tag:blogger.com,1999:blog-19092890.post5297014873643220432..comments2024-03-28T17:41:43.970+02:00Comments on Image Sensors World: Low Light Video DenoisingVladimir Koifmanhttp://www.blogger.com/profile/01800020176563544699noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-19092890.post-33583545214522808432022-06-20T18:44:00.174+03:002022-06-20T18:44:00.174+03:00The paper is accepted at CVPR 2022 as an oral pres...The paper is accepted at CVPR 2022 as an oral presentation and will be presented this week. [Friday June 24, 2022]Atul Inglehttps://www.blogger.com/profile/17531107862096234519noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-36426171445976941482022-06-06T19:48:35.943+03:002022-06-06T19:48:35.943+03:00These image noise is dominate by VFPN, which is ha...These image noise is dominate by VFPN, which is hard to see on current CIS. I wonder the power it took to fix FPN and Hnoise will increase dramatically. magskyichttps://www.blogger.com/profile/01863338961197365344noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-80408687841709170612022-05-18T23:16:49.752+03:002022-05-18T23:16:49.752+03:00BTW, the project page is https://kristinamonakhova...BTW, the project page is https://kristinamonakhova.com/starlight_denoising/ and the direct arxiv link is https://arxiv.org/abs/2204.04210. It is CVPR 2022 paper, not clear if accepted or just submitted so far. There is one more thing in the paper that is a bit confusing: The night videos are shot at only 5-10 FPS, but the paper states the exposure times are 0.2 to 0.1 ms. I think this must be a mistake - it must be 200ms to 100ms. I see substantial motion blur for the fast moving things in the videos. Still the results are really cool IMO.Tobi Delbruckhttps://www.blogger.com/profile/11104032358783859871noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-28913230165096803412022-05-18T23:11:33.280+03:002022-05-18T23:11:33.280+03:00A skeptic like me would say that yes, they do clea...A skeptic like me would say that yes, they do clearly show perceptually superior denoising by comparison to other methods, but the lower light level is nearly entirely due to the Canon high QE and huge pixels that are about 100 times the area of economical MP automotive or surveillance CIS. The superior denoising might be mostly due to very specialized attack on the read and shot noise and column FPN of this particular CIS. And the computational costs are missing, so it's not clear if the method is practical. I hope they improve their published paper with these aspects.Tobi Delbruckhttps://www.blogger.com/profile/11104032358783859871noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-18832143182530563452022-05-18T14:16:06.247+03:002022-05-18T14:16:06.247+03:00They also use a super expensive Canon CIS with 19u...They also use a super expensive Canon CIS with 19um pixels and RGBN Bayer pattern... this chip is about 2MP but measures at least 40x20mm, it is HUGE. Plus I'm somewhat skeptical of their fancy noise modeling - it's not clear it would work in mass production, would each camera need it's own fine tuned denoiser? And it's not clear from their experiments how well the method uses Eric Warrent's ideas of "smart spatiotemporal summation" to reduce noise and motion blur. After reading the paper, I have no idea if this method could be practical to implement in real time, would it need a 200W GPU attached to the camera?Tobi Delbruckhttps://www.blogger.com/profile/11104032358783859871noreply@blogger.comtag:blogger.com,1999:blog-19092890.post-63550720840059551972022-05-18T14:09:55.307+03:002022-05-18T14:09:55.307+03:00This is a cool preprint - it could use a little mo...This is a cool preprint - it could use a little more TLC to make it really nice. I wish these CV people would include some cost estimates for their DNNs... they seem to think the only FOM that matters is accuracy, and everything else, like energy, speed, memory, don't matter at all.Tobi Delbruckhttps://www.blogger.com/profile/11104032358783859871noreply@blogger.com