I just watched the tutorial on creating a video panorama. The process is straightforward enough, but the premise of the whole exercise is not clear at all, to me anyway.
The tutorial starts out with 800 or so PNGs - these pictures represent the raw materials, so to speak. But where are these pictures coming from? How have they been generated? Were they taken manually? It seems unlikely. Was a video camera mounted on a tripod and simply rotated 360 degrees? If so, can I then somehow extract individual pictures from the movie?
Also, for this exercise to work, must the pictures (i.e. the raw materials) be based on a 360 degree rotation or could they also reflect a linear movement?
Thank you very much for your help,
Franco
Creating a video Panorama - please clarify
-
Hopki
- Gnome
- Posts: 13025
- Joined: Thu Jan 10, 2008 3:16 pm
- Location: Layer de la Haye, Essex UK
- Contact:
HI Franco,
The images were taken and stitched in to Equire's before they come into Pano2VR.
So in the tutorial Point Grey suppled the files from their 360 Camera.
Another resource to look at is Here.
Regards,
Hopki
The images were taken and stitched in to Equire's before they come into Pano2VR.
So in the tutorial Point Grey suppled the files from their 360 Camera.
Another resource to look at is Here.
Regards,
Hopki
Garden Gnome Support
If you send an e-mail to support please send a link to the forum post for reference.
support@ggnome.com
https://ggnome.com/wiki/documentation/
If you send an e-mail to support please send a link to the forum post for reference.
support@ggnome.com
https://ggnome.com/wiki/documentation/
-
JimWatters
- Posts: 329
- Joined: Thu Oct 07, 2010 6:16 pm
- Location: Saint John, New Brunswick
- Contact:
Frames for Panoramic video are usually created from multisensor camera rigs. Rigs that have many cameras to capture a panorama all at once.
Some rigs produce spherical images directly out of the camera others need to be stitched together first.
The Theta is such a camera with two lenses and sensors that produce a full spherical image in one press of the button. The camera can be controlled by a smart phone and capture a pano about every 5 seconds.
http://theta360.com/
Others have put many GoPro Hero cameras together to create panorama rigs that can shoot video.
http://freedom360.us/product-category/mounts/
http://shop.360heros.com/360Heros-360-V ... s/1817.htm
Each frame of these videos need to be stitched together to make a series of panoramas using software like PTGui and VideoStitch
http://ptgui.com/
http://video-stitch.com/
The tutorial starts at this point.
Some rigs produce spherical images directly out of the camera others need to be stitched together first.
The Theta is such a camera with two lenses and sensors that produce a full spherical image in one press of the button. The camera can be controlled by a smart phone and capture a pano about every 5 seconds.
http://theta360.com/
Others have put many GoPro Hero cameras together to create panorama rigs that can shoot video.
http://freedom360.us/product-category/mounts/
http://shop.360heros.com/360Heros-360-V ... s/1817.htm
Each frame of these videos need to be stitched together to make a series of panoramas using software like PTGui and VideoStitch
http://ptgui.com/
http://video-stitch.com/
The tutorial starts at this point.
- Jim Watters
http://photocreations.ca
http://photocreations.ca
Guys, thank you very much for your help. Let me see if I understand.
For my project I am thinking of getting 3 to 4 video cameras, arranged on some kind of rig. Taken together, they would cover 360 degrees. The view up and down is not very important in this case, so no need to produce a spheric panorama.
I would then like to attach the rig on a quadrocopter. The quad would then fly along a predefined path with the cameras shooting video.
From such a run I would get three (or 4) video clips. Now, If I understand you right, I would have to extract from each video clip the individual frames. For example, 30 sec à 15 fps times 3 = 1350 individual frames, i.e. 1350 pictures.
I would then use pano2VR as presented in the tutorial mentioned above to generate a single panorama video.
Is that how it would work?
-Franco
For my project I am thinking of getting 3 to 4 video cameras, arranged on some kind of rig. Taken together, they would cover 360 degrees. The view up and down is not very important in this case, so no need to produce a spheric panorama.
I would then like to attach the rig on a quadrocopter. The quad would then fly along a predefined path with the cameras shooting video.
From such a run I would get three (or 4) video clips. Now, If I understand you right, I would have to extract from each video clip the individual frames. For example, 30 sec à 15 fps times 3 = 1350 individual frames, i.e. 1350 pictures.
I would then use pano2VR as presented in the tutorial mentioned above to generate a single panorama video.
Is that how it would work?
-Franco
-
JimWatters
- Posts: 329
- Joined: Thu Oct 07, 2010 6:16 pm
- Location: Saint John, New Brunswick
- Contact:
After extracting the frames from the video and before processing the panos in Pano2VR the frames need to be stitched into panos.
PTGui has a batch stitcher that can stitch a folder of panos using a template.
VideoStitch can stitch the raw footage together without having to extract the frames first also using a template created in PTGui.
Check out this introduction video to VideoStitch.
http://youtu.be/chnqU-ory3g
PTGui has a batch stitcher that can stitch a folder of panos using a template.
VideoStitch can stitch the raw footage together without having to extract the frames first also using a template created in PTGui.
Check out this introduction video to VideoStitch.
http://youtu.be/chnqU-ory3g
- Jim Watters
http://photocreations.ca
http://photocreations.ca
Hi!
Pano2VR actually creates/builds the player to play the interactive video.
The video itself has to be created with other softwares.
First you have to register the video, which you already said will be registered with four different cameras. Now those four different video images have to be stitched to only one video to be played by the Pano2VR player.
To do so, first you have to create an image sequence from the four individual cameras.
You can do that with QuickTime Pro, as used in the tutorial. In the case of the tutorial Karyn does it with the already stitched panorama as she used as an example a stitched one. You are not there yet.
Once you have all the individual frames of the four videos, you have now to stitch them. That means that at the end you will have as much panoramas as frames from the camera.
For this step you have to use PtGui and its batch stitcher, which will stitch all the individual images/frames from the 4 cameras and create the panoramas. In the batch stitcher you have to use “Generate new projects” with the “Multiple
panoramas per folder with a fixed number of images”. In your case 4. Of course that to use the batch method first you have to create a template with one set of individual frames. Frame one from camera 1, 2, 3 and 4, save as template and apply it later in the batch mode.
These panoramas will be the frames from your final and panorama video.
Actually what you got at this point is the “image sequence” which is “Step one” from the tutorial video you are refering to in your original post.
From here on follow the indications from the tutorial. Start at 00:24.
Regards, Martin
Pano2VR actually creates/builds the player to play the interactive video.
The video itself has to be created with other softwares.
First you have to register the video, which you already said will be registered with four different cameras. Now those four different video images have to be stitched to only one video to be played by the Pano2VR player.
To do so, first you have to create an image sequence from the four individual cameras.
You can do that with QuickTime Pro, as used in the tutorial. In the case of the tutorial Karyn does it with the already stitched panorama as she used as an example a stitched one. You are not there yet.
Once you have all the individual frames of the four videos, you have now to stitch them. That means that at the end you will have as much panoramas as frames from the camera.
For this step you have to use PtGui and its batch stitcher, which will stitch all the individual images/frames from the 4 cameras and create the panoramas. In the batch stitcher you have to use “Generate new projects” with the “Multiple
panoramas per folder with a fixed number of images”. In your case 4. Of course that to use the batch method first you have to create a template with one set of individual frames. Frame one from camera 1, 2, 3 and 4, save as template and apply it later in the batch mode.
These panoramas will be the frames from your final and panorama video.
Actually what you got at this point is the “image sequence” which is “Step one” from the tutorial video you are refering to in your original post.
From here on follow the indications from the tutorial. Start at 00:24.
Regards, Martin
Ok.Pano2VR actually creates/builds the player to play the interactive video.
I guess by "register" you mean "shoot", right?First you have to register the video, which you already said will be registered with four different cameras. Now those four different video images have to be stitched to only one video to be played by the Pano2VR player.
By "image sequence" do you mean an image sequence for each of the 4 videos I have shot?To do so, first you have to create an image sequence from the four individual cameras.
I didn't know I could use QuickTime Pro 7 for that. Thank you for this information. That's handy.
Things are beginning to be understandable - until now the whole thing looked like black magic to me.That means that at the end you will have as much panoramas as frames from the camera.
So, if have have 4 movie clips, and each movie clips consists of 900 frames, PTGui will have to stitch 900 panos, is that it?For this step you have to use PtGui and its batch stitcher, which will stitch all the individual images/frames from the 4 cameras and create the panoramas
Question 1: once pano2VR has created the player, can I then also control how the video pano is played using the controls available in the media player, i.e. play, pause, look left look right etc.?
Question 2: Can I also connect several video panoramas using pano2VR pro?
Thank you very much for your help, this saved me a lot of money for additional software I really don't need.
-Franco
Hi!
First yes, by register I mean shoot.
Yes an image sequence for each of the 4 videos. This image sequence will let you build panos later. In QuickTime Pro 7 choosing in Export “Movie to Image Sequence” create a sequence for each camera.
If you have 900 frames in each clip you have to build 900 panos. Easy done with the PtGui Batch Stitcher.
This step is not in the tutorial because they already have the panorama stitched as they create an image sequence from an already stitched video. Your are a step before, but the process is the same.
You can add buttons in the skin as with any other skin. Now I don´t know regarding the hotspots inside the panorama itself, because it moves. Sure you can add in the skin buttons or links to other videos.
Remember that the output video is 3x2 therefore you have to limit the up and down view as shown in the tutorial.
Good luck!
First yes, by register I mean shoot.
Yes an image sequence for each of the 4 videos. This image sequence will let you build panos later. In QuickTime Pro 7 choosing in Export “Movie to Image Sequence” create a sequence for each camera.
If you have 900 frames in each clip you have to build 900 panos. Easy done with the PtGui Batch Stitcher.
This step is not in the tutorial because they already have the panorama stitched as they create an image sequence from an already stitched video. Your are a step before, but the process is the same.
You can add buttons in the skin as with any other skin. Now I don´t know regarding the hotspots inside the panorama itself, because it moves. Sure you can add in the skin buttons or links to other videos.
Remember that the output video is 3x2 therefore you have to limit the up and down view as shown in the tutorial.
Good luck!
Hi there,
thank you very much this clarification. I have two more questions, however (should be the last ones!)
a. "Limit the view the up and down view as the movie is in 3x2 format": not sure what this specifically mean. Is there some specific aspect ratio that is more suitable than others when shooting the individual movie clips ?
b. The tutorial seems to indicate that the final result must be played in Flash. Pleeeeease tell me that ain't true!!!
-Franco
thank you very much this clarification. I have two more questions, however (should be the last ones!)
a. "Limit the view the up and down view as the movie is in 3x2 format": not sure what this specifically mean. Is there some specific aspect ratio that is more suitable than others when shooting the individual movie clips ?
b. The tutorial seems to indicate that the final result must be played in Flash. Pleeeeease tell me that ain't true!!!
-Franco
The original sample video used in the tutorial is a equirectangular image therefore is has a proportion of 2x1. but once transformed it has a 3x2, therefore there will be black holes at the top and bottom of the image. To avoid to be seen, the vertical viewing parameters have to be reduced. Watch 4:50 in the tutorial. In your case if you shoot with 4 cameras, maybe anyway you won´t get an equirectangular output once stitched.
That means that you won´t get a 180º vertical view. It all depends of the cameras you will use and its position. You have to test it.
Sorry but it is true...the final result must be played in flash. At the moment.
Regards!
That means that you won´t get a 180º vertical view. It all depends of the cameras you will use and its position. You have to test it.
Sorry but it is true...the final result must be played in flash. At the moment.
Regards!
180 degrees vertical FOV wasn't in the plan anyway, so that's not an issue for me.
The Flash thing - now that is indeed a show stopper, since eventually the videos have to be played on mobile devices, and I can't ignore the iPad can I?
That is not for tomorrow, but it is indeed a very important requirement moving forward. Now, developers don't often make promises as to when a new feature will be included, so I fully understand if you are not willing or able to make any promises. If that's the case I would appreciate it if we could have a (brief) bilateral and confidential conversation (Skype?), since this feature is really quite important to me. I am going to have several business meetings with potential customers in the next month, and being able to even mention panorama videos would be a very strong and unique selling point.
Best,
Franco
The Flash thing - now that is indeed a show stopper, since eventually the videos have to be played on mobile devices, and I can't ignore the iPad can I?
That is not for tomorrow, but it is indeed a very important requirement moving forward. Now, developers don't often make promises as to when a new feature will be included, so I fully understand if you are not willing or able to make any promises. If that's the case I would appreciate it if we could have a (brief) bilateral and confidential conversation (Skype?), since this feature is really quite important to me. I am going to have several business meetings with potential customers in the next month, and being able to even mention panorama videos would be a very strong and unique selling point.
Best,
Franco