Client needs indoor drone video, would the Air 3 extra sensors help over the 2s?

[WildcatDave]

Hey guys and gals. I currently have an Air 2 that is going to get replaced in the next week or two with an Air 2s or Mavic 3. I don't know if I really need the M3 as I mostly shoot outdoor commercial buildings and real estate and the extra sensors are unnecessary for that work. I also shoot in very good daylight at base ISO100, so the 4/3 vs 1 inch is unlikely to matter much.

I wanted to move up to higher megapixels, 20MP vs 12MP. I know my existing Air 2 has a 48 megapixel mode and using it definitely does bring out quite a bit of extra detail. However there sometimes are strange artifacts (bayer pattern?) on smooth surfaces like concrete or in the sky and it creates additional noise. Also I don't need files that large, it's just more to work with than necessary (RAW shooter). Additionally, the larger sensors do create an overall better image than my existing Air 2, which has a tiny sensor. I shoot Full Frame on my DSLRs. The zoom lens on the M3 is a total novelty and something I've never even thought I would need so it's probably unnecessary, and its tied to a tiny sensor like the Air 2.

The question relates to a client that needs quite a few video shoots of both exterior and interior (the interior are very large gyms). I'm wondering since I probably won't have GPS communication indoors consistently if the extra sensors will save me from some possible trouble. Most of the shooting will be live during the day with people participating in normal gym activities. So it's going to be a sweaty hands stressful work. The first round of shoots will pay for either drone so it's not really a money issue. But I don't see a reason to burn an extra $1,000 or more after additional batteries and such. Also if I get the 2s I can use my existing batteries, case and accesories from my Air 2. Though I may pick up an extra case and use the old Air 2 as a backup, especially since I'm going out of state for this shoot.

Opinions and thoughts from everyone welcome. If you have shot indoors in ATTI (non GPS) mode, please share any tips. I did let the client know I will need some time to get the new drone and get myself squared away. Also a good friend of mine owns the small local gym that I go to and he will give me as much time there as I need to do practice runs.

I attached a sensor size comparison image. Most of you may know that the "1-inch" sensor in the Air 2s is not actually 1 inch, neither is the 4/3 sensor in the M3 actually 4/3 of an inch, as they are based on some old video standards. I think it's a bunch of BS that they use that, it's a marketing ploy. However, you get what you get, and all the companies do it. And the Air 2 sensor is so tiny in comparison it's still a big jump up.

 

[Eagle Eye 62]

When you say Air 3 I assume you mean Mavic 3. Without getting into a technical minutia discussion I would say the Mavic 3 will give the best image quality but may not be the best choice to fly indoors due to the size and lack of availability of prop guards. I have a Mavic 3 and a new Mavic Mini3 Pro. I just ordered some prop guards for it. Since it is very small and offers excellent image quality, it might be the best option for your indoor application.

 

[WildcatDave]

When you say Air 3 I assume you mean Mavic 3. Without getting into a technical minutia discussion I would say the Mavic 3 will give the best image quality but may not be the best choice to fly indoors due to the size and lack of availability of prop guards. I have a Mavic 3 and a new Mavic Mini3 Pro. I just ordered some prop guards for it. Since it is very small and offers excellent image quality, it might be the best option for your indoor application.

Yeah I corrected that post. I forgot it doesn't have "Air" in the name. The mini might be a good option indoors since the lighting will be good but I'd really prefer to get a larger sensor otherwise I can just use my Air 2 and not spend any money. The slight size difference between the Air 2/2s and Mini 3 Pro and even the mavic 3, it's just not that significant. Better indoor flight is what I'm most concerned about and the extra sensors on the mavic 3 might help for that. And I'd also like to get a better quality sensor if I'm going to spend some money and the Mini 3 Pro sensor is decent but not really much of a step up. Or maybe it is, and I haven't seen much since its so new. Oh jeez, I thought I had it narrowed down to the Mavic 3 or Air 2s...

 

[WildcatDave]

When you say Air 3 I assume you mean Mavic 3. Without getting into a technical minutia discussion I would say the Mavic 3 will give the best image quality but may not be the best choice to fly indoors due to the size and lack of availability of prop guards. I have a Mavic 3 and a new Mavic Mini3 Pro. I just ordered some prop guards for it. Since it is very small and offers excellent image quality, it might be the best option for your indoor application.

I noticed Amazon has a DJI branded guard for the mavic 3 now, Reviews are not great and its $50. 3rd party ones should start showing up soon...

Still no ND filters for the Mini 3 pro. Which is a bit strange considering it's large aperture, it's really going to need them. Find it odd that DJI didn't release it with at least its own branded ones for the Fly More Combo. Looks like there are some third-party ones. It's another expense to add in.

 

[SoCal2S]

It's a tough choice based on your description of the use cases. It doesn't seem like image quality will be a huge differentiator for that particular line of work, but that can always change with a new client. If money is no issue, I think the M3 will be better due to the side collision avoidance sensors for indoor work. So it depends on your level of skill and risk tolerance, is it really worth the extra money? Especially when accounting for your existing batteries that are compatible with the A2S.

Based on the fact that you said money isn't an issue, go with the M3, because you'll always be scratching that upgrade itch if you don't.

 

[WildcatDave]

It's a tough choice based on your description of the use cases. It doesn't seem like image quality will be a huge differentiator for that particular line of work, but that can always change with a new client. If money is no issue, I think the M3 will be better due to the side collision avoidance sensors for indoor work. So it depends on your level of skill and risk tolerance, is it really worth the extra money? Especially when accounting for your existing batteries that are compatible with the A2S.

Based on the fact that you said money isn't an issue, go with the M3, because you'll always be scratching that upgrade itch if you don't.

This was my pattern of thought as well. It was mentioned that the Mini 3 might be easier to fly indoors because it's smaller and the quality of footage is pretty impressive. I can't deny either of those thoughts. I likely should pick up a Mini 3 and a Mavic 3 and sell the Air 2. I might just start with the Mini 3 and see what I think... Too many options!

 

[spsphotos]

Side note, if you are flying indoors make sure you put on prop guards, especially if you are flying near people which is sounds like you are. I'd also get more comfortable with ATTI and realize that you need to make sure the drone does not drift (it will keep going even if you let go of the sticks). Extra sensors on the A2s can't hurt.

 

[WildcatDave]

Side note, if you are flying indoors make sure you put on prop guards, especially if you are flying near people which is sounds like you are. I'd also get more comfortable with ATTI and realize that you need to make sure the drone does not drift (it will keep going even if you let go of the sticks). Extra sensors on the A2s can't hurt.

For sure, prop guards will be here tomorrow. What if the drone gets a few GPS links, will it still work? What if they are weak, can it go IN and OUT of ATTI mode? Does it stay in ATTI once it switches over until next flight? The A2s has no more sensors than my current A2 or the A3p I am also considering. The Mavic 3 has the extra side sensors, and I know I will be making strafing and curving routes. I will probably know better once I test out my current A2 at my gym.

Will be starting practice today...

 

[brett8883]

Here’s a few things to keep in mind.

My experience with this same type of work is if you can find a way to avoid using a drone indoors you should. Would putting your DSLR on a pole get the same perspective? It would be much better quality and would avoid all the issues flying the drone indoors.

With my experience trying this I personally would not try flying a drone indoors with people around. That would be a hard no for me especially not with anything larger than a mini. Even with a mini you really need to try to find prop guards that are more like a cage design like the mini 2 prop guards. Haven’t seen similar for the mini 3 yet but maybe they exist?

Another issue with drones indoors is the lenses really aren’t wide enough. I think they are like 24mm full frame equivalents which is good for outdoors but not ideal indoors. You are a real estate photographer, how often do you use 24mm on your DSLR indoors?

I don’t think the sensors are active if the drone is in atti mode which in all likelihood will be the case flying indoors so I don’t think that matters.

Even with good indoor lighting, these sensors are so small, even the mavic 3 that you will have to use high ISO indoors. They are designed for outdoors in full sun which is many many times brighter than any indoor setting.

Try to find a way to do it with your big camera if you can. That’s the best advice I can give you.

 

[MAvic_South_Oz]

What if the drone gets a few GPS links, will it still work? What if they are weak, can it go IN and OUT of ATTI mode? Does it stay in ATTI once it switches over until next flight?

This is why with any drone, it would be beneficial to have the option the simply fly in atti if desired.
Dropping in and out of gps mode, positioning mode with vps ( if low enough) is fraught with dangtfor a pilot indoors.
No, if you re battery up or switch off for a new flight it will go through the process of working out the flight mode (gps, positioning, atti) and on it goes with remaining vigilant with what’s happening during a flight.
You can turn vps sensors off and eliminate that happening at least.

Edit - went and double checked, found a DJI forum thread that says you can’t turn vision sensors off with the Fly app ?

 

[WildcatDave]

Here’s a few things to keep in mind.

My experience with this same type of work is if you can find a way to avoid using a drone indoors you should. Would putting your DSLR on a pole get the same perspective? It would be much better quality and would avoid all the issues flying the drone indoors.

With my experience trying this I personally would not try flying a drone indoors with people around. That would be a hard no for me especially not with anything larger than a mini. Even with a mini you really need to try to find prop guards that are more like a cage design like the mini 2 prop guards. Haven’t seen similar for the mini 3 yet but maybe they exist?

Another issue with drones indoors is the lenses really aren’t wide enough. I think they are like 24mm full frame equivalents which is good for outdoors but not ideal indoors. You are a real estate photographer, how often do you use 24mm on your DSLR indoors?

I don’t think the sensors are active if the drone is in atti mode which in all likelihood will be the case flying indoors so I don’t think that matters.

Even with good indoor lighting, these sensors are so small, even the mavic 3 that you will have to use high ISO indoors. They are designed for outdoors in full sun which is many many times brighter than any indoor setting.

Try to find a way to do it with your big camera if you can. That’s the best advice I can give you.

Trust me I don't WANT to fly indoors. That said, it is explicitly what the client wants. They have previous videos done this way and they did some remodels and want updated videos. The person they used before did very good work, but was very expensive. I have been doing their remodel photos and virtual tours and they asked me if I could do the drone work. They also said if I did not want to do it, that was not a problem. However, I don't want them to (in the future) look for an all-in-one solution. It is for a VERY big client (Gold's Gym) and I would like to offer them all the services. This is the break I have been working at for many years. I understand there are some considerable risks, but to step up you often have to to step out of your comfort zone.

I have thought about using my very capable Panasonic m4/3 gear to do the video work, but they want similar videos and you can tell it is drone footage. At least they are lit very well, I don't think I would have to go above ISO400 to shoot 4k/30fps at 1/60 shutter. The gyms are huge with 8-10 meter ceilings, so some of the work will be a bit easier. That said some of the current videos show flight paths into very tight areas (like the Sauna). Honestly I think its a bit overkill, but the client is boss. Here is a link to the most recent video they had done so people can better evaluate:

I have some time to practice at my local gym to see if this is something I can do. If I don't feel comfortable I will back out. But that's not how I've made it this far as an independent business owner for 30+ years. So my mentality is if others can do it, so can I...

 

[MAvic_South_Oz]

Many of those shots could be have with even a phone camera gimbal, or Pocket / Pocket 2, the hallways, sauna, no need for a drone, and I'd be really surprised if one was used in those ground level shots.
It really looked like gimbal hand camera in that sauna, definitely no sensors on if a drone, and who knows how many takes for a good steady clip.

The rest ?
Sub 250g drone, full prop cage (mini 2 ?), turn off all sensors, I'd be surprised if you had GPS inside there, but possible time to time . . . full agreement for any people present disclosure (even then you might be on shaky ground legally) . . . it might be legal by FAA indoors to do the OOP ? Not sure, but liability wise, open there if anything happens most likely, even with agreements.

Many of the scenes were obviously carefully posed / staged, much like those incredible tiny whoop scenes that are famous for fly throughs of bowling alleys, cafes, bars etc.

 

[brett8883]

Trust me I don't WANT to fly indoors. That said, it is explicitly what the client wants. They have previous videos done this way and they did some remodels and want updated videos. The person they used before did very good work, but was very expensive. I have been doing their remodel photos and virtual tours and they asked me if I could do the drone work. They also said if I did not want to do it, that was not a problem. However, I don't want them to (in the future) look for an all-in-one solution. It is for a VERY big client (Gold's Gym) and I would like to offer them all the services. This is the break I have been working at for many years. I understand there are some considerable risks, but to step up you often have to to step out of your comfort zone.

I have thought about using my very capable Panasonic m4/3 gear to do the video work, but they want similar videos and you can tell it is drone footage. At least they are lit very well, I don't think I would have to go above ISO400 to shoot 4k/30fps at 1/60 shutter. The gyms are huge with 8-10 meter ceilings, so some of the work will be a bit easier. That said some of the current videos show flight paths into very tight areas (like the Sauna). Honestly I think its a bit overkill, but the client is boss. Here is a link to the most recent video they had done so people can better evaluate:

I have some time to practice at my local gym to see if this is something I can do. If I don't feel comfortable I will back out. But that's not how I've made it this far as an independent business owner for 30+ years. So my mentality is if others can do it, so can I...

I recently had like exactly the same thing happen to me so I get it. I was asked to do indoor drone of a gym, this one was vacant though, and I told them I could do it.

I bought a Mavic Mini 2 and a DJI FPV not knowing which one would be better. I thought maybe the first person view from the FPV would make it easier to navigate but I really didn’t end up even trying it indoors (its scary powerful). I flew the mini 2 around but I just felt like I didn’t have enough control to get into tight spaces and the lens just wasn’t wide enough most of the time. You can’t be scared of crashing which is why you need a mini 2 or mini 3 (if they have prop cages for the mini 3.) I think it has a bright lens too (f/1.7 instead of f/2.8). I think you might be underestimating how much light is in there. Our eyes are very good at adjusting to low light.

I ended up refilming everything with my Sony A1 on my RS 2 gimbal and just made camera my movements look like they were from a drone and that worked so much better. I peppered in some of the actual drone footage and honestly you can’t even tell which footage is from the drone and which is from the gimbal. The client loved it and when I told them it was mostly from a gimbal they said they didn’t even know such a thing existed and didn’t care because what I gave them was exactly what they were looking for when they asked for “drone footage.”

Moral of the story is that I think a lot of the time when people say “drone footage” they just mean stabilized flowing camera movements. My advice would be to get some select footage from the drone but use a gimbal for the rest. In that video you linked above 99.9% of those shots could have been with a gimbal and I’m not convinced they weren’t. If a drone was flying that close to people they would be looking at it and would be scared of getting hit. I really think you need a mini for this. An Air2s or Mavic 3 are just too big and the stakes much higher if you hit someone.

My video below.

 

[TheRock]

Yea, I was going to say that too. Most of them look like gimbal shots and are well-staged. Some of those you can do 360 camera on a pole too. Most are short so not that hard to pull off. A very good camera was used for this initially. If this is indeed a client you hope to retain, get the Air 2S at minimum or the Mavic 3 to do it right. If it's too tight, go Mini 3. But like brett says above, gimbal is the way to go. Get a sturdy pole.

 

[brett8883]

Actually you can forsure tell a gimbal was used because when they walk up the stairs at the beginning you can see the gimbal shake from walking up the stairs. A drone wouldn’t do that.

 

[WildcatDave]

This is why with any drone, it would be beneficial to have the option the simply fly in atti if desired.
Dropping in and out of gps mode, positioning mode with vps ( if low enough) is fraught with dangtfor a pilot indoors.
No, if you re battery up or switch off for a new flight it will go through the process of working out the flight mode (gps, positioning, atti) and on it goes with remaining vigilant with what’s happening during a flight.
You can turn vps sensors off and eliminate that happening at least.

Edit - went and double checked, found a DJI forum thread that says you can’t turn vision sensors off with the Fly app ?

Thanks for checking on that. Can't believe they removed that ability it's ridiculous. I'm sure people were crashing in that mode and they were doing too many warranty replacements so they just took it out. Does anyone know if the ATTI hack using the software still works since it's an older mavic 2.

 

[WildcatDave]

I recently had like exactly the same thing happen to me so I get it. I was asked to do indoor drone of a gym, this one was vacant though, and I told them I could do it.

I bought a Mavic Mini 2 and a DJI FPV not knowing which one would be better. I thought maybe the first person view from the FPV would make it easier to navigate but I really didn’t end up even trying it indoors (its scary powerful). I flew the mini 2 around but I just felt like I didn’t have enough control to get into tight spaces and the lens just wasn’t wide enough most of the time. You can’t be scared of crashing which is why you need a mini 2 or mini 3 (if they have prop cages for the mini 3.) I think it has a bright lens too (f/1.7 instead of f/2.8). I think you might be underestimating how much light is in there. Our eyes are very good at adjusting to low light.

I ended up refilming everything with my Sony A1 on my RS 2 gimbal and just made camera my movements look like they were from a drone and that worked so much better. I peppered in some of the actual drone footage and honestly you can’t even tell which footage is from the drone and which is from the gimbal. The client loved it and when I told them it was mostly from a gimbal they said they didn’t even know such a thing existed and didn’t care because what I gave them was exactly what they were looking for when they asked for “drone footage.”

Moral of the story is that I think a lot of the time when people say “drone footage” they just mean stabilized flowing camera movements. My advice would be to get some select footage from the drone but use a gimbal for the rest. In that video you linked above 99.9% of those shots could have been with a gimbal and I’m not convinced they weren’t. If a drone was flying that close to people they would be looking at it and would be scared of getting hit. I really think you need a mini for this. An Air2s or Mavic 3 are just too big and the stakes much higher if you hit someone.

My video below.

Thank you for this detailed response. I'm going to watch the videos again carefully. I have a very nice powered gimbal for my Panasonic micro four thirds gear. I've only used it a few times as I was doing real estate video walkthroughs with it and quickly switch to virtual tours which are much more enjoyable for everyone.

For these gyms I'm doing both still photos and virtual tours and now they want videos, so maybe I need to revisit the gimbal and the usefulness of that. When looking at the video I can see where a lot of that could have been done with a gimbal and not needing to fly a drone. Your comment about people noticing the drone and looking at it is all true. I don't know how anyone could not stare at a very loud set of propellers flying near them.

 

[WildcatDave]

Yea, I was going to say that too. Most of them look like gimbal shots and are well-staged. Some of those you can do 360 camera on a pole too. Most are short so not that hard to pull off. A very good camera was used for this initially. If this is indeed a client you hope to retain, get the Air 2S at minimum or the Mavic 3 to do it right. If it's too tight, go Mini 3. But like brett says above, gimbal is the way to go. Get a sturdy pole.

Sounds like from those of you that know a lot of this was done with Kimball so I will attempt to get better at that I do have the RICO theta Z1 360 camera which is how I did the virtuals. Depending on how urgent they need to do this I might be able to get myself in a position to use this equipment and only the drone for the exterior flybys.

I really don't want to attempt to fly a drone inside and judging from what you guys are noticing in the video it looks like that was possibly done very little if at all. I don't know how they would have taken on that risk as their insurance wouldn't cover that if the drone pilot clipped off an ear air with some propellers.

 

[R.Cowan]

I too am looking for advice on flying my Mavic 3 indoors. I'm in the process of providing bi-weekly construction progression video and photos of a 850K square ft warehouse facility on 55 acres. The client is requesting a fly through tour of the facility once it's complete. I have no experience flying indoors and I'm aware that without GPS, I'll basically be flying in atti mode. Will my sensors on the Mavic 3 help me from hitting a wall? I'm a few months out before this phase and want to be prepared.

 

[brett8883]

I too am looking for advice on flying my Mavic 3 indoors. I'm in the process of providing bi-weekly construction progression video and photos of a 850K square ft warehouse facility on 55 acres. The client is requesting a fly through tour of the facility once it's complete. I have no experience flying indoors and I'm aware that without GPS, I'll basically be flying in atti mode. Will my sensors on the Mavic 3 help me from hitting a wall? I'm a few months out before this phase and want to be prepared.

Not in atti mode. If you are close enough to the ground and enough light for vision positioning the OA sensors should work but not in atti mode.

In an 850k sqft warehouse that actually might be fun especially if they don’t have really tall shelving throughout the core. That’s where that DJI FPV drone might excel.

You might have more issues with low light though. Those warehouse buildings are usually pretty dimly lit. Usually they’ll have a lift at most warehouses and you could see if they’ll let you get up on one of those and fly from there so you can get a better view of the drone while flying around.