Capture One 22 Panoramic Stitching and HDR Merge

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
[Music] hi everyone my name is walter rowe and i'm here to demonstrate capture 122 new pano stitching and hdr merge features in this picture we have a panorama which was comprised of two rows of images actually 17 images that were stitched together if i go back to the browser you can see if i select all the nef files these are nikon z7 raw files they're lossless compressed raw files you can see that i took the first row of images that's actually the water line itself and then i took the second row of images which captures the sky there's overlap between the first row and the second row with the trees and there's overlap within each row so that the piano stitch is able to both vertically and horizontally stitch together all these images all i've done is just select the whole batch of images and this is going to take a while i'm going to give you a warning in advance and i may talk through this while it goes and and i may show you the resulting image that i did earlier but here we're going to right click and we're going to say stitch to panorama and it's going to bring up a dialogue that's going to offer us a preview in the pano stitch dialog there are multiple choices and you can see from the little icon adjacent to each choice what source images should be should would work best with each algorithm spherical images you can tell it has multiple rows of images the cylindrical you can tell it's just a single row of images perspective would be where you actually have a converging point in the center you can see how it's larger on the sides and smaller in the middle so if you're doing architecture or perhaps you're looking down a long street in a city where you have anything where you have converging lines in the middle and then panini is where you might have converging lines but you also may have a featured subject in the middle so we're going to wait okay so here we now have our preview now notice i've currently got selected cylindrical it remembers the last item that you had selected when you did a panorama stitch previously so the last time i did a panorama stitch i had used cylindrical so that's what it already had selected when i brought this up i'm going to switch to spherical and we're going to see that it's going to regenerate the preview for spherical this is when you have multiple rows of images that you're stitching together one thing to note which was quite clever on the part of the developers for capture 122 is that if you do switch from one projection to another it caches the result of the selections uh the the projections that you've already have previously selected so if i go to cylindrical it's remembering that the cache from the from when i had selected it before when i go to back to spherical it remembers that the cached image so it doesn't have to regenerate that all over again now if i choose perspective since i haven't chosen that before for the sequence of images it will have to generate that preview so we'll have to sit through it okay so here we have a cylindrical view and we have um spherical view now spherical actually does look better in my opinion relative to what i actually saw this actually stretches it cylindrical it actually stretches it a bit too tall spherical the tree line actually isn't super tall in real in reality um you can see there's a black block in the bottom left corner and the bottom right corner i did tell you that this was 17 images it's an odd number so i i probably took eight images across the top or nine images across the top and eight images across the bottom and that probably explains it why it doesn't really fill in all the way to the ends but what i actually find impressive is it figured that out it didn't matter that i had a different number of images across the bottom row than the top row it managed to figure out how they all stitched together in which ones belonged on the top versus the bottom now it would be nice if we could zoom in right here in the preview and really see how well this was done unfortunately we can't do that we also can't choose to go in and deselect pictures from here for example if i wanted to get rid of one picture because i just felt like i didn't need these on the end there's no way to deselect that one other item of note in the stitch is you can choose between 175 50 and 25 so um i don't necessarily need to have a 200 megapixel output from this so i could choose a smaller dimension i'm going to choose the 200 megapixel the 100 in this case notice it's 27 980 pixels wide by 7480 pixels tall i'm choosing that just because i want to be able to zoom in to a full zoom for you so that you can see the result of this stitch and how clear it is so now we're going to watch it it's going to go through this will not take 30 seconds this will take considerably longer this this could take minutes i'm running on a 2017 uh imac with 32 gigs of memory i think it's a radeon graphics card i do have acceleration enabled this will take some time to generate this i have done this previously and i'll just show you this this image right here is from a previous run of this stitch and while this is actually stitching you can see it's actually nicely showing us the progress that as it goes across um but i'm gonna go ahead and zoom in to 100 uh for this image and you can see uh this these were all um i think these were handheld i can't remember for certain if i had a tripod but this was definitely not a super scientific you know carefully laid out sequence of images i literally just went to this lake and decided it was going to capture the set of images and i did you can see that it's very very clean now there might be a little artifact across here that might represent maybe a little sign of the of the horizontal i'm sorry the vertical stitching between the two rows which i had not actually noticed before but i mean quite frankly the detail is amazing this is a 47 megapixel camera i can zoom right across here uh there's really just no artifacts even in the water reflections they're just no artifacts of the stitching together of the images as i pan across here it's just remarkable it's beautiful it stitched together well i mean it's a very it's a very detailed image and in my opinion a very detailed image might offer lots of opportunity for smudging uh as an artifact uh to show when um when you know where where the images had been stitched together uh the one thing that i will say here is um you notice that there's a fisherman here who's got his back to me um i maintain the focus on the far side of the um of the lake because the trees were really the focal point but i thought having the fishermen in the picture was uh just a nice uh kind of gave it up a sense of scale and and place and and all that and right here you can see right here in the fishing rod that maybe it's missing a little bit of the fishing rod i don't know if he moved the fishing rod and so it had a hard time finding something there it's entirely possible he was not perfectly stationary but quite frankly since it's out of focus and we're looking at a hundred percent if i zoom back to fifty percent it's not hardly noticeable if i zoom out to uh you know fill the screen um view you really can't tell it at all it's kind of just hidden in the trees over there so you really can't tell if i were to even print those pretty large you'd probably have to look for that to even notice it you probably wouldn't even pay attention to it because you'd really be looking at the trees so as i said there was maybe this little artifact here um this line here uh that might be where the two rows may be stitched together a little bit but i'm not sure um that could have actually been some artifact in the cloud itself but it's certainly not visible anywhere else now let's go back to the browser now here's the resulting image that we just produced so you can see that it that it's not cropped it's got the little um jack you know the the edges so we'll have to go in a crop i'll set an unconstrained crop so i can uh i can get the uh you know crop it to it to what i want i'll pull this in past here to eliminate the side little black blocks in the corners i'll pull up from the bottom just to chop off the the pieces there i have given capture one feedback that it would be nice to have an auto crop mechanism where maybe it could actually detect where it needs to crop um so here now would be effectively the image and if i go and look at this version of the image let me go back and look at this cloud again if i go back and look at this version of the image that we just produced that horizontal line that was in the other version is no longer present and i will tell you that i was a beta tester and it's conceivable that i generated that other version with the last beta release and not the final production release so actually i think this is even now improved i'm gonna go back here just for for speed sake i'm to copy the adjustments from from the old one and paste them on the new one uh just so that we get the the kind of like for like and actually looks like you notice if i go to that where it says pano multi i think i use the cylindrical projection for that one and this one is the is the spherical projection which is the multi-row projection and i think it actually did the right thing i think that might have been the difference in why we don't have that artifact in the clouds um i think that probably made a difference so i've learned something new here uh let's take a look at the fisherman let's come back over here 100 percent uh no we still have this kind of little missing piece of the fishing pole it's still there but i think you you i think this demonstrates well enough um how amazing this stitch is like i said it was an odd number of images it was not an even number of images it wasn't like nine over nine it was nine over eight and it found you know which row had more versus view and it pieced them together perfectly so i'm really very impressed with the result that they have done with this pano stitching just super impressed and of course you can see how beautiful the the image is now i also did just a single row and for this i used some really old pictures i had um these images are from 2006 you can see 2006 in the file name these are really old pictures of a building uh that i actually used to work in many many years ago and you can see that's just you know horizontal pictures a sequence of landscape oriented pictures and then this was the resulting frame that came out of that so here i will highlight those i will pano stitch them together these are pretty old and small um i'm gonna select cylindrical since it's a single row of images and not a multiple row multiple rows of images and here it stitches them together there's some artifacts in here um there's some some spots and stuff but uh what i'm going to show you with this that impressed me that absolutely the most out of this is how well um this roof line actually came together um so here we go okay so here's the let me go and copy uh my my dust spots layer i want to copy that over here just for let me just copy all the adjustments over here just so we go go quickly in there okay so all this is just dust spots i'll turn on the cube just so you can see where they are the i turned on the the heel just just so you can see the little arrow keys i'll hit the m so you can see all the red dots you can see where they are i do this because i want you to see that there's none on the building itself so i'm going to now i'm going to turn all that back off and we're going to go into 100 view and i'm going to slowly pan across this image and i want you to see see all these really um all these lines in the building all these all the the roof lines from the tin roof they're absolutely perfectly aligned the overlap of the images is absolutely flawless every single detail comes out in this picture this is absolutely uh unparalleled i i have done the same stitch in lightroom and lightroom always for some reason has a smudge right around here where it joins the pictures together and it follows all the way down through the roof line so in my opinion this is my own personal opinion uh let me hope i see another dust spot here i forgot let me hit that um in my opinion this is absolutely better quality than lightroom when it comes to pano stitching i've done a few in lightroom and i just think this beats the pants off of what lightroom does for pano stitching and i just couldn't be happier especially with these images that are from 2006 i mean these are just these are old like they may be i think it's maybe a six megapixel image um so stitching them together uh to get this kind of detail was was you know wonderful opportunity i'm glad i took them this way long ago as a as a potential future stitch um i just can't i just i can't say enough good things about this so you've got a multi-row that we've done and we've got a single row that we've done so these are pano stitching so that's how easy it is to do the pano stitching in in the new capture 122 highly recommended can't be beat now let's look at the high dynamic range merge and oops let me go here okay so this was a sequence of images um uh let me just find one that's kind of closer to like normal uh you can see that the scene here we're at a lakefront uh this is a commercial building i think the bottom has maybe a i don't know a spa or something and the top has actually has a whole foods in it and then what you can't see off to the right is some restaurants and things and it's just a pretty a pretty setting in a planned community so what i did was i merged this together uh using just five of the frames i selected every other image in the sequence so i just selected you can see five of 14 selected um so i just did that when i did this merge and what i found was selecting um all uh let me go back here if i select all nine of the bracketed sequence it did not improve the quality it did not improve the quality in any way shape or form in fact the recommendation from capture one directly is that you really only need three images and you should have a metered exposure and a minus two stops exposure and a plus two stops exposure and the rationale that i'm being told is the the sensors today modern sensors have such an enormous dynamic range that that's really all you need in order to get effectively an extra two stops of of information on either side so you've increased your dynamic range by four um by doing a plus two and a minus two from your metered exposure now of course you need to do your own exposure compensation uh based on what the scene is itself so don't just take a metered exposure at zero compensation make sure you set the exposure comp for what you think is true to the scene we always have to learn how to use the meters in our cameras so here we'll just uh select these five frames um i am using five instead of three but that's that's my choice that's what i'm choosing to do um in fact you know what here let's try this let's get rid of the the center ones let's go for the the extremes so we've got a minus i don't even know what the difference is between these honestly let me look at the information and see what it says let's look at these and we'll find out so we've got 1 100 of a second we got 1 400 of a second so that is two stops and we got one fifteenth of a second so you've got one fiftieth one 125th so this might be a little more than two stops difference on this one uh so let's go and select that one instead it's 1 125th okay so we've got a 1 100 we got one 125th and one 400th so we got some metered exposure some minus two and plus two and let's see what we get i'm going to right click and here we're going to say merge to hdr we're going to get a different pop-up for this one here we're going to get auto-adjust and auto-align you can turn off auto adjust and it'll make no adjustments whatsoever but i think you do want to use auto align this especially if you're hand holding images where you might have a little a little shake i have a nikon z7 and the z7 has the ability to take a bracketed sequence automatically so i set it up on the tripod and just told the camera to take them so i know that they're they're aligned quite well but i'm letting it do auto align anyway and i'm going to let it do auto adjusted also so we'll start this process this doesn't take too long because it's only three images the more images you include the more you will see the you know of course the longer it takes because it has to merge them together but this shouldn't take too terribly long since this is only three images here we go uh hdr so there's our final product and it's been auto adjusted uh so what do we have we actually have um nice clean we have no uh no blowouts let me do exposure warning so here are a tiny bit of exposure uh problems here and if i mouse over it actually may go to a hundred percent let me mouse over you can see they're actually um not quite 255 255 255 and the reason they're showing exposure warning is if i go to my preferences and say exposure i'm actually telling it to warn me at 253. let me put this all the way to 0 and 255 and you'll see they're almost gone they're practically gone because there is a 254 there might even be a 255 in there somewhere uh so you know that just tells you this you know this did a really remarkable job and that's all well and good let me go back and close those um so uh i think this looks pretty good white balance might need to be adjusted i don't know i might want to pull in a little bit of uh let me turn off exposure warning so i'll get rid of the the red um but you can see that even the shadows actually there's a you know you can see under here it's pretty dark but not totally dark go back over to here and if i want to pump that up a little bit i could i'll add i'll put it to 20 you can just watch the histogram here it does does come right up to the end it does clip a tad um if i really wanted to get uh crazy about it i could actually come into um oh look at here oh that's all because auto levels let me reset the auto levels and then we won't have that see now we don't have that problem we do probably want to have a little bit of level in here come up just because we want to get the brightness that we want but i don't need to clip on the darks the darks actually look pretty good uh so yep that looks pretty good um i might want to get a little more contrast i might want to get a little bit of clarity so i don't push the contrast too far maybe a little structure for for a little more detail sharpening um but i think this looks really really good i'm quite uh quite happy with this and again this was uh by selecting the uh a metered exposure and then a plus two and a minus two uh from the from the metered exposure and i think this turned out quite good um so well done and let's go back and now let's let's look at some some extreme examples a buddy of mine provided me these examples because they are indeed pretty extreme so here's a one one sixtieth all the way to one twenty five hundredth of a second and then there's a one-sixth 40. so if 640 is our baseline and then we say what that's 320 160 that's minus two stops and then 640 12 80 2500 that's two stops so we've kind of got the we'll call this one the metered and we'll call this the plus two and this is the minus two um perhaps his camera takes the over uh the mean under and then the meter at the end but if we you know if we uh blend these together and merge the hdr and we'll just pop that up uh i'm actually you know you can see this is a pretty pretty good example test because this is bright light sunshine this is like middle of the day and there's there is auto adjust on here and then i did my own you know addition of adjustments uh you can see this is this is one eye this is where i did it before and just did some adjustments you can see there's the adjustments are different um but i mean this looks this doesn't look too bad it needs it needs some fixing let me just apply the adjustments now that now i've got the same adjustments uh from the from the other version that i've applied to it and i'm telling you for a midday shot this is pretty impressive let me go to the camera info and see this was taken at 2 30 in the afternoon in june uh and just so you know this is in in the northern hemisphere this is straight up summer uh right at the peak of brightness in in the afternoon and it's really phenomenal what what this is actually pulled out so i'm i'm quite pleased with that result and here's another example in this example this is a real estate example so you've got a kitchen it's got some nice french doors that go out to and outside in the corner you've got this little windows that has bright sunlight coming in of course you've got mixed lighting because you've got you know indoor lights have a color temperature outdoor has a different one in the underexposed frame you can see there's some detail outside and of course the others it's almost entirely blown out and so we're going to merge this together merge to hdr go and let this run and uh in the resulting image that i did previously now here you can see this has been auto adjusted and so forth and and maybe the auto adjustments weren't great um you can see some detail outside so that's nice you know you can but it's not the best it still needs some work um so i went back and did my own adjustments on the the image myself actually i think this is even a bit blue blue there we go uh and then i don't know if it's maybe a little too red even also so i'm gonna copy the adjustments i'm gonna go to this image i'm gonna undo all the auto adjust and apply the images that i had the adjustments i had before looks like it's a bit underexposed so let's just we'll add one stop pretty close yep so uh i think this is quite nice um if you're doing high-end real estate you probably wouldn't be satisfied with this a really high-end market really wants like all of this outside to be perfectly visible you're going to use some other tool probably you're probably not going to use capture one but for you know certainly for me uh you know i i don't do commercial real estate as a as a business um for me if i were just maybe on a on a tour somewhere and i wanted to take a an exposure bracketed sequence just to have a memory really well it's really very acceptable in my opinion and then the next sequence is uh from a a commercial laundry mat actually um so this is one-tenth 180th and one over 1.3 seconds so you can see this is really a challenging you know kind of low super not well lit inside and very bright sunlight outside and just quickly i'll do these just just so you see it but i'm not going to spend a lot of time on it but i just really wanted you to see you know there are some kind of these are kind of some extreme examples uh these are from olympus cameras these are um uh i think they're olympus micro four thirds so here's uh here's the result that came out just now so not too bad this has auto adjustments on it um i think i did some adjustments earlier on a rendition of it but i think this actually turned out pretty good straight out of the box with auto adjust um again maybe if you're doing high-end maybe you're not super happy but i mean the details of what are outside isn't that important it got all the details inside and you know if you're if you're trying to get a a lease on a laundromat or something you know probably probably would work fine so that's the new uh pano stitching and high dynamic range for capture 122 my name is walter rowe if you want to purchase capture one and want to use the links in my description of this video i would really appreciate that and uh i look forward to seeing you next time make sure you click on subscribe so you get notices when i do other videos or when i do a live stream thanks you
Info
Channel: Walter Rowe Photography
Views: 352
Rating: undefined out of 5
Keywords: Capture One, Photography, How-To, Raw Editor, Tutorial, Post Processing, Digital Asset Management, Capture One Sessions, Capture One Catalogs, Live Stream, Panoramic Stitching, High Dynamic Range Merging, HDR Merge
Id: ncSu_ejYnNw
Channel Id: undefined
Length: 32min 29sec (1949 seconds)
Published: Mon Dec 13 2021
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.