Hello, my name is Michael. I’ve been part of OSM for almost 20 years and have been gaining experience in creating 360° images for Panoramax for over two years.
I thought it was now time to put my experiences down in writing.
My wiki editing and translation skills aren’t particularly strong, so perhaps someone could help improve my attempt.
There are a number of other points to add, particularly in the section on mounts and tripods.
The graphics and images of settings in applications mostly contain German text. It will take a while before I have created adequate English versions.
Everyone is warmly invited to discuss, add to and improve the content.
Also about nadir… I shared 2 command lines to done it, starting from a standard PNG, reprojecting it and applying it without JPEG recompression (preserving the original picture quality)
Reproject the nadir PNG: convert panoramax-logo.png -rotate 180 -distort DePolar 0 -flip -flop -geometry 5760x576! -sampling-factor 2x1 nadir-5760.jpg
Apply it on your pictures: jpegtran -copy all -optimize -trim -drop +0+2304 nadir-5760.jpg original.jpg > nouvelle.jpg
Dimensions are valid for Gopro Max 360 pictures resolution (5760x2880) and need to be adapted for other resolutions.
We plan to add nadir automatically in a near future on Panoramax upload API.
Thank you all very much for your replies!
It seems there are a number of ways to resolve the issue with the nadir.
I completely agree with you.
Actually I don’t fill areas with a single colour; I only did so in the Wiki to illustrate the process.
I usually use patterns that I create from an area near the nadir. This works quite well, but if the type of road surface changes, it looks less elegant. It’s nothing dramatic, but if there were better solutions… The possibility of creating a pattern that incorporates elements from several images in a series is very interesting. Thank you very much for the script.
My journey for cover the nadir took me from initial hardware solutions…
…and simple patches with text on them…
…through to very understated designs…
…and those linked to an event with custom text.
Throughout the entire process I learnt a lot about GIMP, its capabilities and how to use it. And although designing the nadir cover – beyond simply filling it with colour – is purely a matter of GIMP expertise, I think I’ll describe these methods in the wiki as well.
Ideally, I’d like to be able to edit the desired area of each image in the same way as you can e.g. with the Heal Selection Tool in GIMP. Or perhaps something using AI capabilities, who knows…
But I think that, for now, the easy-to-use tools we have at our disposal are sufficient to achieve good results for a variety of needs.
I really like mounting the camera directly on the helmet. It’s simple and easy and in most cases removes the need to add a nadir in post-processing.
One quick PSA for this kind of mount: if your helmet has MIPS, make sure that the way you attach your camera doesn’t lock the MIPS part to the helmet. (In a crash, you want the MIPS section to move freely to absorb some of the impact.)
As I don’t have a helmet to hand, I haven’t had any experience with this.
However, I think it’s a very simple and very common way of mounting a camera.
Perhaps we should compile the various aspects of this type of mounting and add them to the wiki as well.
I’d be happy to add the relevant information and media to the wiki.
I did some testing with a helmet mounted 360 camera. To get the camera high enough I used a gooseneck. It was very hard to get it stable on the helmet, especially the sideways stability was a problem. I used tiewraps to fix the helmet mount tight enough, but then the whole helmet turned sidewayson my head, because of the high placed weight. I had to fill the room in the helmet completely (used some baseball caps) and pulled the helmet straps really really tight to make it stable enough… only to find that when hiking, the camera had to be lowered very often because of the vegetation and other low passages.
In short, my experiences were not that great, I would not recommend a helmet mount. Even if you don’t mind being shunned by everyone you meet, because you look ridiculous, walking with a helmet and gooseneck.
Edit:
If you mount the camera directly on the helmet, it’s much more stable, but you always have yourself in the picture instead of what you walk on. For my purpose, documenting hiking routes, this ws not acceptable, I wanted a much smaller visible ground circle, meaning: higher camera position.
Keeping the camera horizontal is a real issue when mounting it on an helmet… but we are testing Panoflat, a bot that computes pitch and roll on 360° pictures and modifies metadata to have them displayed correctly.
I have a very basic question: what’s a general rule for how much storage space one should have on an SD card for a set amount of time of recording images? I know this will vary with the resolution of the camera, but having some baseline information from your GoPro Max would be helpful! How many GB per hour at 2 second intervals vs 5 second intervals?
Fresh from the field…
Today, about 70 minutes of recording time on the bike.
Settings: 360° time-lapse, 2-second interval.
2,144 images, total 7.1 GB.
I’m using a 256 GB card here, which I reformat two or three times a year.
You can get by quite comfortably with a smaller card too.
BTW: Starting with 100%, after these 70 minutes, the battery still had 56% capacity remaining.
vielen Dank für deinen Erfahrungsbericht und die Arbeit an der Wiki-Seite. Gerade praktische Hinweise zu Ausrüstung, Befestigungen, Aufnahmeabständen und typischen Fehlerquellen sind für Panoramax sehr hilfreich.
Ich habe ergänzend ein kleines Tool erstellt:
Das Foto Geo-Tool soll Bilder und kurze Videosequenzen für den Panoramax-Upload vorbereiten. Hintergrund ist, dass viele Apps und Messenger – zum Beispiel WhatsApp – Metadaten wie GPS-Informationen, Zeitstempel oder andere EXIF-Daten entfernen oder verändern. Für Panoramax sind diese Informationen aber sehr wichtig.
Das Tool soll dabei helfen,
vorhandene GPS-/EXIF-Daten sichtbar zu machen,
fehlende oder fehlerhafte Metadaten zu erkennen,
Bilder und kurze Videosequenzen für die weitere Verarbeitung vorzubereiten,
Bildserien besser zu prüfen,
mögliche Probleme vor dem Upload zu finden.
Wichtig: Der letzte Schritt funktioniert derzeit noch nicht. Die vom Tool vorbereiteten Bilder werden von Panoramax aktuell noch nicht akzeptiert. Die genaue Ursache ist mir im Moment noch nicht bekannt.
Ich habe den Ersteller bzw. Ansprechpartner von Panoramax dazu bereits per E-Mail kontaktiert, bisher aber noch keine Antwort erhalten.
Das Tool ist deshalb momentan vor allem als Prüf- und Vorbereitungshilfe gedacht. Für einen vollständigen Panoramax-Workflow muss noch geklärt werden, warum Panoramax die vorbereiteten Bilder nicht annimmt und welche technischen Anforderungen dafür genau erfüllt sein müssen.
Vielleicht wäre ein Hinweis auf solche vorbereitenden Werkzeuge in der Wiki-Seite sinnvoll, zum Beispiel in einem Abschnitt „Vorbereitung der Bilder“ oder „Prüfung von Metadaten vor dem Upload“.
Das Tool ist noch in Entwicklung. Rückmeldungen, Tests und Verbesserungsvorschläge sind willkommen.
Grüße aus Thüringen
gabischatz
I’ve finally found the time to post a tutorial on this in the wiki.
I’m no GIMP expert, so I’m sure there are plenty of ways to do this better and more easily. But it works. If you want to have a go at things like this, it’s not a bad idea to have a look at the GIMP manual every now and then. You can discover lots of interesting things there.
The possibility of creating a pattern that incorporates elements from several images in a series is very interesting. Thank you very much for the script.
