here’s a shorter version of the december 2015 time-lapse video i posted the other day..

i was asked by Udi of diyphotography.net to write about how this was made so i will explore some of it below. i did write a bit about the technique and idea the other day when i posted the long version of the time-lapse. you can check the film and writing out at: markmarano.com/2016/01/17/december-2015-time-lapses-exploring-new-perspectives-aerial-stereographic-projection-little-planet-infrared-still-in-motion-driving-walking-from-above-etc/

most of the shots in both films were created with either the ricoh theta m15 or the ricoh theta s camera. the shots over or near water (and a couple of the ones on top of the car) with the larger, more noticeable black circle in the middle (surrounded by a couple of faded out / reflected portions of the black circle) were done with the ricoh theta m15 camera inside the ricoh th-1 hard case. for those unfamiliar, the case is not completely waterproof though provides protection against splashing and debris, though as you’ll see in the shots with it, it does lower the image quality and if you watch the longer version of the video, you’ll see in the beginning of the shot on the rocks that it caused a large reflection. this probably would have been a lot less noticeable or perhaps even disappeared altogether if the thin side of the camera was pointed at the sun. though the light was different in the very first little planet shot, if you look at the small white piece of the camera in the center of the frame, you’ll see that it was closer to being positioned like this — the combination of the sun being partially behind the clouds and the positioning of the camera relative to the sun made a big difference as the reflection seen of the th-1 hard case is much less noticeable and yields a much more pleasant shot.

the cameras were mounted in various ways for the different shots. if you look closely at some of the shots, you’ll notice a white piece surrounding the little bit of the camera that is visible in the center of the little planet. this is a piece of 3/4″ pvc pipe that is stuck into the sand, dirt, or water. it’s around 2 feet long cut from a longer piece with a cap at the end that has had a 1/4″ hole drilled in it for a 1/4″ threaded screw that the camera can mount to. you can attach the camera directly to the 1/4″ screw though, in attempt to protect the plastic tripod mount on the camera from breaking (as i had this happen on my original ricoh theta camera), i’ve been keeping a double head stud (1/4″ to 1/4″) made by smallrig with a thick rubber washer on the bottom of the camera — i placed a 1/4″ coupling nut on the bolt/screw on the pvc pipe (or whatever pole, tripod, or other mounting rig i’m using) so that this 1/4″ stud can thread into it. with the small double head stud attached to the bottom of either camera, it still fits in the original gray case (that came with the ricoh theta or ricoh theta m15 camera). for the shots by the water with the camera inside the th-1 hard case, i mounted the th-1 hard case directly to the pvc pipe in the ground, or for the shot on the rocks, i carefully set the th-1 case directly on top of one of the rocks as i think i recall it being difficult to insert the pvc pipe into the ground there with additional rocks under the sand.

if you watch the longer version of the video, the camera starts to move at the end of the first little planet shot — i simply picked up the pvc pipe and held it over my head as i walked for this part of the shot. i did the same for one of the shorter shots near the beginning where i was walking in nature (except that the camera wasn’t in the th-1 case for this one and you can see me in the shot as i wasn’t hidden beneath the case). if you don’t want to show up in the shot, a simple way to block yourself is to place a rubber fender washer between the camera and the top of the pole and be careful to keep the camera directly over your head. with a taller pole it becomes simpler for you to disappear as your further distance from the camera will make you smaller and easier to hide under the part that is blocked by the black rubber washer. with a tall enough pole you may not even need the rubber washer. i have a 15 foot long konig & meyer 23770 microphone fishing pole that i’ve held up directly over my head with the the camera where you don’t see me, or if the pole isn’t centered, i’m so small in the shot that i’m barely noticeable. in this video, i used that pole (in conjunction with two 20″ and one 12″ matthews microgrip 3/8″ rods) for the shots of the walkway over the highway (for both the night and day shots) though it was resting on sidewalk rather than over my head to reduce swaying. at first i had also added one 20″ and one 12″ microgrip rod with the 1/4″ threads though i didn’t even get to fully extend the 15′ k&m pole as the whole thing started to bend and i didn’t want to risk something snapping and falling down onto the highway below. to see some shots of me holding the pole over my head you can watch the “winter and spring 2015 time-lapses” video i did last year — i think all of the quick shots from overhead in the beginning of the video were done with that pole (without the microgrip rods.. perhaps one or two shots somewhere in that video had a 1/4″ threaded rod as an extension though i can’t remember for certain at this point — most if not all shots were just with pole and it provides a nice height and perspective on its own). that time-lapse film also has shots with the theta mounted on a dji phantom quadcopter (before the plastic tripod mount broke). you can see that time-lapse video here: markmarano.com/2015/10/22/winter-and-spring-2015-time-lapses-aerial-stereographic-projection-little-planet-infrared-etc/

for the shot of me walking (wearing a hat in the center of the little planet near homes and tall palm trees by the water), the camera was mounted on top of a 123.5″ manfrotto 142b high stand extension pole (with an additional 20″ or 32″ of matthews microgrip rods i believe). the bottom of the extension pole was attached to an avenger avf300 junior wall plate which was screwed into a block of wood. at the bottom of the block of wood, i attached a shepherd 9795 8″ medium duty pneumatic rigid caster so i could roll the whole pole over whatever terrain i was walking on. this setup works well on something paved or hard ground — when walking with the pole in softer sand it became quite difficult to push the caster and i had to lift up the pole and somewhat carry it to not get stuck. a thicker wheel on the caster would help here (which i had expected when i ordered it though the one that arrived was thinner than pictured). most of the shot in this time-lapse that was done with me walking with this pole setup was done using the 1920×1080 video mode of the ricoh theta s from which i extracted 1 still frame per each second of video (using a bash script to run ffmpeg at the command line). the beginning of the shot (seen very briefly in the long version of this video) was done with still images — which gives a better image quality however the minimum time between shots on the ricoh theta s is 8 seconds and that wouldn’t have given enough frames at the rate i was walking with the pole for the nice motion you see with the extracted video frames. i was also experimenting with a custom python script i had written to send the command to take a shot with dr compensation one after another (as the ricoh theta s does not currently support interval timer shots with dr compensation turned on). however this script was running on an old android phone (as i recall the free python app i downloaded for my iphone was an older build that didn’t support all the commands needed to communicate with the camera), and as i had locked the screen on the phone when putting it in my pocket, the python script was not taking shots one after another and there were very long pauses between the shots — it seemed that either android os or the python app itself didn’t support running fully in the background, so i switched to video mode for the walk back (after discovering it had stopped taking shots in the middle) which would also give me more frames to work with. it’ll be really nice if the next firmware update of the ricoh theta s supports intervals as short as 1/2 second or 1-2 seconds between shots (even if this means the images need to be stitched later into equirectangular form on the computer as is the case with the videos). if ricoh doesn’t update the firmware to allow this, the recently announced nikon keymission 360 camera will be a nice option for shots like these as it will support shooting video at 4k resolution which would yield much nicer looking little planets than hd resolution. i’ll have more shots walking with the theta m15 on this pole when i put together the shots i did this past november.

the first driving shot, where the car appears much larger in the center of the little planet, was done with the ricoh theta m15 camera inside the th-1 hard case which was mounted to the moonroof of the car via a small panavise 809 window suction cup mount with a 20″ matthews microgrip rod (and i believe i had a desmond dtr-1 tripod thread reducer with an additional 3/8″ to 1/4″ tripod thread adapter to give more support under the th-1 hard case — though the tripod thread is metal on the bottom of the th-1 case, after breaking the plastic mount on my original ricoh theta camera i didn’t want to take any chances with damaging the case when it was in motion). i drove slowly during this shot as it was on the interval timer mode (the minimum 5 seconds the theta m15 allows). i think most of the shot was under 15 mph as that was the speed limit on the causeway. fortunately there wasn’t a lot of traffic wanting to go quicker for this shot. this shot was successful though the car was larger than i’d like in the shot. i had shot a time-lapse with the same panavise 809 suction cup and i believe just a 6″ extension (without the th-1 case) back near the end of 2014 (that i never rendered or completed) and the car was even larger. the only way to make the car smaller would be to lift the camera up higher above the car though with the 20″ extension and the th-1 case, the whole rod and camera were shaking a little. as the panavise is made of plastic, it probably wouldn’t be a good idea to drive faster or put an additional extension rod on that setup so i began researching what could be done to raise the camera higher. it didn’t seem like anyone had done what i wanted to do before so i had to work with what little semi-related information was available and make some educated guesses (and trust and rely upon my intuition / divine guidance). i ended up with a setup using 3 large 6″ vacuum cups mounted to the metal roof of the car (the car may have been a little more centered in the shot if i placed the main vacuum cup on the moonroof though i didn’t want the larger rig bouncing on the glass as i was driving). for the main vacuum cup i used an avenger f1100 as it was one of the least expensive options available and had a fixed baby pin. on this vacuum cup i attached a manfrotto 099b 35″ to 92″ lightstand extension. the f1100 is not rated for nearly as much weight as other 6″ vacuum cups however i calculated the whole setup on it would be within its recommended weight, though as i would be extending the camera at least 3′ above the car i got 2 additional vacuum cups (made by woods powr-grip) for triangulation. the woods powr-grip vacuum cups had a 3/8″ threaded stud on them to attach standard tripod heads and adapters to. i attached a giottos mh-1304 pro series ii socket & ball head (with fender washers as spacers) to each of these vacuum cups with a 1/4″ to 3/8″ reducing coupling nut attached to 36″ long 3/8″ threaded rods. the mh-1304 was one of the least expensive options i could find that support a decent weight and where the round piece that normally sits under the camera can be removed which would allow me to mount the coupling nut further onto the threads of the mh-1304 for additional support. at the top end of the threaded rods i had giottos and manfrotto ball heads, one attached to a manfrotto 035 super clamp and the other to a flashpoint nano clamp. as the super clamp was heavier and larger, i attached it a little lower than the nano clamp on the 099b pole. in the first couple of shots i used microgrip rods rather than the 36″ threaded rod though for just a few dollars more the 36″ threaded rod from the hardware store seemed like better support with less components to twist off. the only downside is that the rods are a little greasy when handling (though nothing some gaffer’s tape can’t solve).

i experimented with various heights on the 099b extension rod. most shots were done with one of the extensions somewhere in the middle and the other closed (or both just a little bit for slightly more rigid support as i think i recall it wobbling more when just the smaller tube was fully extended and especially with th-1 case at the end in the first shots). for the last shots when driving around honeymoon island state park around sunset and dusk, i had raised the camera higher as i was driving slower — i didn’t measure any of the actual heights though for this one the extension rod was around 6′ long and for most the others it was at least 4′ or 5′ long. paying attention to overhead clearance is obviously an important thing here — the height of the rig alone on top of the car was taller than the height of the car itself (which measures 60″ tall). in one of the parks the camera or pole hit a low hanging branch (despite steering to avoid most branches that appeared low). fortunately i was going slow and no damage was incurred though know that this type of shot requires extra consideration and more awareness while driving. the first shots with this setup i drove slow in a nature preserve with no traffic. as i tested the setup and saw it was okay and holding, i slowly started going a little faster with it and making sure it’d be fine. if attempting to create a shot like this, its important to check the vacuum cups every so often, especially before you know if your particular rig is working properly. on occasion, one of the rear vacuum cups needed to be re-pumped as the red line started to show. if you’re wanting to compare measurements to your car to do similar shots, the car is a 2015 honda fit ex (manual) — measuring only 160″ long and 67″ wide, it shows up nicely in the center of the little planet with the rig. the car was actually pretty dirty though due to its size and lack of resolution in the shot (and perhaps being white orchid pearl in color), it still appeared bright and clean. regarding cleanliness, its important to clean the area where the vacuum cups are being mounted (and make sure the vacuum or suction cup is clean of debris too). i brought water and paper towels to clean the area where i was mounting the cups to before every shot. with this setup, i was able to drive up to 45-50 mph. when the camera was higher for part of the one evening, i probably drove under 20-25mph most of the time and at most 30-35mph for a little bit on the causeway before pulling over to park. the traffic was backed up (i had never seen so many people there and leaving the park all at once) and it ended up giving some nice long exposure shots when i was parked waiting for it to clear up. for most of the shots with this rig, i ran with the camera outside of the th-1 case. it seemed that the th-1 case was causing more vibration probably due to its larger size and additional resistance to the wind, even at low speeds. i probably also had the pole extended just using the thinnest part of it when i tested the th-1 on the rig which would have been less support and more likely to move. as my ricoh theta s camera had a defective lens (a large part of the image was out of focus) and was going to have to go back to ricoh anyhow, i didn’t worry as much about needing the protection of the th-1 case when driving. luckily there were no incidents, not even bugs smacking up against the camera, and it was fine. it was nicer to not have the th-1 case or the desmond dtr-1 tripod reducer for extra protection as this allowed for the rig to take less visual space in the little planet. you’ll notice in the first couple shots how much more of the car is blocked by the th-1 or the dtr-1. with a black car it wouldn’t be as big of a deal, or you could photoshop every single frame to remove the rig, though i prefer to do everything in camera and leave the computer work simply for the “developing” rather than retouching or cleaning up stuff that could’ve been done during production and creation of the shot.

when i developed the mounting method, i realized that i’d also be able to use a lot of the gear to be able to mount the camera off the front of the car to get another unique perspective of the car driving on the little planet. this mounting method was more common and close to how i believe others create car images (such as what i’ve read is done in drifting photos and likely some car advertisements where the background is blurred but the car is sharp and in motion). i placed the two woods powr-grip vacuum cups on the side of the car on the panel above the front driver’s side tire. i was going to use the avenger f1100 but it didn’t hold with the curvature of the car in those locations. with a manfrotto 186 mounting adapter, i placed the 035 super clamp on rear vacuum cup and i used an old manfrotto mini ball head to mount the nano clamp onto the front cup. there was around a foot of space between the ends of both vacuum cups. the 099b pole was held by the clamps and fully extended with the ricoh theta s attached via a smallrig 3/8″ to 1/4″ brass tripod thread reducer. again, i started off driving slowly in a more controlled environment before taking the setup out on the roads. in the first shot, the camera was lower to the ground though after i saw how you couldn’t see much of the windshield, i decided to adjust the angle for the next shots. with this setup, even on the main country/rural roads, i tried to keep the speed not more than 35-40mph as the pole was swaying a lot (though may have gotten up to 45mph for short distances). i didn’t try the 142b pole, though it probably would’ve swayed less however it’d be additional weight hanging off the car (and probably would be a lot better off being held by two super clamps instead of one nano clamp and a super clamp). with this setup, after at least one of the shots, i saw that both of the vacuum cups had released a little bit and were showing the red lines, however they were still holding on tight to the side of the car. although you don’t need to think about overhead clearance driving like this, you’d likely need to be more careful with the swaying, loss of suction, and also going around turns (as you need to be more aware and extra careful if there’s a car on the other side of the road (or in the next lane over) that the pole could hit or come close to since its sticking out a few feet in front of the car). thinking about the setup as i write this, it may have been a stronger grip to have the larger super clamp up on the front vacuum cup as it could balance the weight a little better than having the nano clamp up front.

regarding the camera settings for the driving shots, you’ll notice in some of the shots the car appears to move more and bounce around the little planet more — these were shot with the video mode where i extracted 1 frame per second from the video recording (after converting the video files that came out of the camera into equirectangular videos, using the ricoh theta computer application). it seems that the software is applying corrections for the camera position when it renders the equirectangular video which causes this. when the time-lapse appears to be going faster and the car is in a fixed position on the little planet frame, these were done as still shots with the interval timer setting. when rendering the stills, i did not apply corrections — after i created the video, i read that these corrections are stored in the exif data of the still images — a script could be created to read the GPano:PosePitchDegrees and GPano:PoseRollDegrees for each image and apply these corrections to the pto files used to render the little planets in the open source hugin software. for at least one or two of the interval timer sequences, i shot it with manual exposure so that the brightness of the little planet wouldn’t vary if the sun was behind the clouds or i was driving under trees or whatnot. if not on the manual mode, i would at least pick the iso priority mode so i could set a manual white balance to prevent colors from shifting between images. for the night shots, i had to experiment with the manual exposure to get a single setting that would work. it was tricky as some roads had brighter lights than others and at first the exposure was a little too long and it appeared like the headlights were overexposing the road at their brightest spot — i ended up using 1.3 seconds at iso 100 for the shots with the rig on the top of the car. still at this setting the car was sometimes overexposed if the shot happened to be under a bright light though overall the look is good. i also experimented with having the fog lights on versus off for some of this shot. in the last driving shot at night where the camera was hanging out in front of the car, i had to drop exposure to 1/2 second at iso 100 so as not to overexpose the area around the headlights too much (yet keep a decent exposure for the lights and background).

the black and white shots were created with a gopro hero4 silver camera that i modified to shoot infrared images. i removed the original lens that came with the camera (there are various techniques out there telling you how to do this — i opted to put the camera in the oven at its minimum temperature to soften the glue and then twist the lens off with pliers so i didn’t have to open up the camera) and i replaced it with a lens that was corrected for distortion and did not have an infrared cut/block filter in place. (be aware you could completely destroy your camera if you attempt to do this and something goes wrong) i recall attempting to glue an IR pass filter to the back of the first lens though it was getting messy and the glue was blocking part of the image area, so i ended up placing the filter in front of the lens and inside the outermost plastic ring that sat around the original lens (as the new lens happened to be shorter than the original fisheye lens when it was installed). this still fit inside the waterproof case or could be used with the official gopro lens protector when outside of the waterproof housing. for the driving infrared shot, i used the smaller gopro bike mount on the 099b pole between both clamps after it was rigged up for the theta s camera. for the aerial shot of the kite and sky, i mounted the camera on a small wooden piece i had devised that hangs right on the kite line (with a rope cleat for it to stay in place on the line). the breeze must’ve been strong as the kite was flying pretty high that day and the camera ended up pointing higher than i wanted to — i had wanted to get some of the ground in the shots (and there’s only a very little bit seen briefly in some frames) — this was also the first time i was mounting the gopro with the distortion free lens (that has a narrower angle field of view) on the kite rig instead of the regular fisheye lens. [a lot of times when you’re trying something new or learning a technique, you may not get any usable images.. don’t let this discourage you if it takes a couple or few tries to get something you like — each time you attempt something new you’ll learn what to do or not to do for the next time when you do get your shot. and its okay to start off simple and just practice with one new technique at a time. i realize i’ve combined many different techniques/looks/ideas in this video though this all came with experimentation and taking it one step at a time (and sometimes lots of research and development and not everything always working out as i liked). each thing you learn is something you can add to your recipe book of ideas. the more you practice and fluent you become with a technique, the simpler it becomes to combine them and take them to the next level, if you feel so inclined or inspired]

for the little planet aerial shot walking on the beach at the park, i mounted the ricoh theta s camera on a brooxes compact picavet cross with brooxes hangups (and the pekabe ball-bearing blocks which make it smoother) that was hanging on a 200 lb line on an into the wind 9 ft levitation delta kite. hanging on the picavet, i had either a 12″ or 20″ microgrip rod so that there’d be more room between the camera and the picavet which would allow me to render a wider angle view of the little planet and sky. [i had shot a couple of time-lapses with the original theta around a year ago where it was mounted directly to the picavet and that limited how the little planet could be rendered, especially as in the shot i completed it was really windy and i recall the angle of the little planet not being straight down from the camera and needing to be compensated for, further reducing the angle of view possible — so i wanted to make sure that i’d have more flexibility with the extension rod.] the kite was a bit of a workout. i planned it with the forecast of the wind so most of the shot i was walking in the direction where the wind was either a little to the side or pulling the kite ahead of me rather than me having to fight the wind — this was a big help compared to the long shot i did a year ago as i was fighting the wind during the first half of that shot. i think i measured the whole walk being 8+ miles and it was a bit longer than anticipated as when i was taking out the extra spars from the kite bag, i took them all out and had to go back to the car as i couldn’t launch the kite when i later realized this. i don’t recall the last time i launched that kite, it may have been over a year. i suppose sometimes it helps to double check things if you haven’t used them or done a certain technique in a while. [slowing down also helps.. not only in becoming more clear about what you’re doing when creating an image and feeling the inspiration, but also in enjoying and fully experiencing the whole process of exploring and creating and being in a beautiful place and discovering its unique and subtle details] when walking the whole island, there were tons of people in the beach areas near the parking lots. i had never seen so many people at the park. the whole new big beach was pretty full and walking around with a 9 foot wide kite with a 24 foot long tail tends to draw some attention. the theta camera was hung lower on the kite line than the gopro camera so it would have a clear view of the ground (besides the kite line that shows up in the shot). i don’t know the exact height as i didn’t measure how much line i let out nor the angle of the line, though from some of the photos, it appears that the horizon was close to the top floor of the nearby condo buildings, meaning the camera was probably only around 60-80 feet up in the air (putting a similar perspective within the reach of really tall pole aerial photography or elevated photography, which i plan on experimenting with sometime in the near future).

the shots of the car driving around the planet were pretty simple. i had thought of doing these a long time ago though never did as it would’ve taken much more time and effort to move the car just a little bit between interval timer shots. with the theta s supporting hd video, i opted to trade off some resolution for simplicity and just started the video and then drove the car continuously in a couple of circles around the camera. for the shot with the camera on the ground of the parking lot, i put together a number of 1/4 to 3/8 adapters and perhaps a washer or two to create a weight that would hold the camera down and avoid the risk of the wind or anything knocking it over onto its lens. i did this rather than use a pocket tripod as the weight was narrower so it wouldn’t show up in the center of the planet (and it was likely a little closer to the ground than a tiny tripod would’ve been).

for one of the shots where it looks like the camera is sitting fairly still in the water with the waves splashing over it, i attached the larger gopro bike / roll bar mount to a 16″ long stake (from a home improvement store). at first the stake wasn’t in the ground enough and it started to wash away into the water. luckily something told me to check or look in that direction (as i was back at the bench where i had tied down the kite to in order to start to take it down). i ran out there and grabbed it before it washed away and stuck it deeper into the ground. drilling a hole to the top of the stake and inserting a 1/4″ or 3/8″ set screw would work better (though i figured i’d experiment with the gopro mounts i already had before deciding what size set screw would work best to give me options to mount the th-1 case or an extension pole to the stake). the wide angle shots near the water were shot with the regular gopro lens though it was “defished” by applying a camera profile correction to the images inside of lightroom.

the less wide angle looking shots in the video were done with my iphone (either holding down the button for continuous shots or taking one by one with hdr turned on). in the longer version of the video there are also some shots that were taken as videos on the iphone and then i converted them to still frames — for the shots of the water going over the rocks, i think i used 5 still frames from each second of video.

as you see, most of the technique for the shots involves creative ways of mounting the camera. the technique itself besides mounting the camera is fairly simple. i think i mentioned earlier the minimum interval timer settings of the the cameras. most of the interval timer shots were done at the minimum setting (either 5 seconds or 8 seconds depending on the theta m15 or theta s). the manual long exposure night shots were done at the minimum interval based on the exposure time — for 30 second shots, the minimum interval is 31 seconds i believe and for 60 second shots, the minimum is 61 seconds — effectively, the camera is continuously shooting for these. most of the long exposure shots were done at iso 100 with either the 30 second or 60 second shutter speed, and probably a couple shots were a little shorter, around 20 or 25 seconds. for the shots with the pole over the highway at night, one of them was with the iso bumped up to 1600 and a 1/25 shutter speed, one shot was a video, and one was with 1 second exposures at iso 100 (which worked out pretty well considering the pole was moving being that high in the air).

the final shot was a shutter speed of 30 seconds at iso 100 and it ran for just over an hour before the fully charged battery died. all the shots in the entire time-lapse film were rendered at 10 frames per second, so for this last shot, the 112 total images exposed play back in 11.2 seconds in the time-lapse video. a lot of other people seem to shoot time-lapses to be rendered at higher frame rates (closer to standard movies or tv at 24-30fps), though i prefer the look of the slower frame rate of 8 to 16 fps [i think part of it is that a slower frame rate has a feel of older film and that plays nicely especially when the footage isn’t completely smooth or if it has a grainy look]. if you notice in this last shot there is no sign of a tripod or anything the camera is mounted on — this is as i stuck a 1/4″ threaded rod directly into the ground and mounted the camera on it [being narrower than the camera itself, it effectively disappeared, yielding a very nice looking little planet. this is really the trick to making a little planet with nothing or nearly no signs of mounting equipment in its center — keep the pole or rod narrower than the camera and if it does need some sort of base, keep it as high as possible. in some of the shots, i used a light stand with extension rods to hold the camera instead of a regular tripod as the legs being closer to the ground are further from the lens and become smaller in the center of the little planet. another thing i’ve begun experimenting with this month is mounting the 099b extension pole with microgrip rods to the super clamp which is attached to a railing — with the super clamp being 12′ away from the camera it begins to visually disappear in the shot]

this is getting quite long going over the details of the various shots so i will begin to wrap it up. for the software side of creating this, i’ve mentioned lightroom, hugin, ffmpeg, the ricoh theta software, and custom bash scripts. these are all you really need to create these. the image color and density corrections are done in lightroom (and in some cases, i never even completed them as i did some quick settings so i could export the files and run some tests, and then i liked the way they looked or got tired of this whole process taking weeks and kept them as they were). in lightroom, i also used the star ratings to keep track of which images would be in the time-lapse and which would be cut out, and i rendered all the files to a temporary folder as tiff files (to maintain maximum image quality). i rendered the images for the little planets in various batches — all the ricoh theta m15 shots were done separate from the ricoh theta s images and those were separate from the ricoh theta s videos. for each of these sets i wrote a custom bash script to generate pto files and run hugin to process all of these into little planets. after each of these first takes, i rendered a video (using ffmpeg) of what i had to get a feel for the look and then customized the script to alter the hfov of the images, and then for some shots, i modified the script to allow it to change the hfov, pitch, roll, and yaw throughout the shot so the little planet was effectively moving, subtly rotating, or changing in some manner. i ended up making mistakes (probably as i was rushing or tired or something) and some of the shots took a few takes to render the way i liked. all of the still images for the little planets were then combined into one folder with the gopro and iphone shots, and then i wrote a script to generate a list of all of the files which i used in order of capture time so i could tell ffmpeg what files to include in the time-lapse and in what order. after rendering it, i ended up going through multiple iterations of rendering to further edit out some frames i didn’t need or add ones back where i had cut the shot too short. the list of files also had to be adjusted in the beginning as sometimes more than one camera was running at once and i had to group the respective shots together so the time-lapse wouldn’t flash back and forth undesirably between the different cameras throughout the entire shot. before i created this shorter version of the time-lapse video, i rendered the full version (that i linked to above) which was simpler to edit as most of the full length of the shots were included with only basic editing out of the beginning and end of shots and a couple/few unwanted frames in the middle.

if you’ve found any of this helpful and will be purchasing any equipment for your own projects (or anything else from amazon), please click the following special link before checking out (you’ll pay the same whether or not you use this link, though if you do click it before checking out amazon will give me a referral credit toward future purchases): amazon.com/?tag=marano-20

thanks and thanks for reading and watching! please share this post or video if you enjoyed it or know someone who will. enjoy your own creative endeavors! and don’t get too caught up in the technical details [most images and time-lapses i create simply come together out of experimentation and exploration]. yes the technical side is needed at times to create unique images though just allow the inspiration to flow through you and listen to your own inner guidance, and you’ll find yourself making great images and experiencing beautiful places. embrace the adventure and journey!