3d Model - Can DD automate the shots required at lower elevations


#1

8 may edis EDIT :slightly_smiling_face:
Ok so i am have added the DroneDeploy labs “beta auto flight modes” which seems to have suggestions for the overlap recommendations at different elevations. of course i imagine it cannot really anticipate the height of scructure and thus for a very close flyby it would seem to need additional oblique. Other edits above . sorry i am master of typos due to voice dictation

Hi All
i am new to DD and have not reviewed all topics and youtube messages ( nor yet got answer from DD support)

I am testing with a Mavic Pro, iOs and DD to create high resolution 3d Models
I have discovered, as seems obvious, that a top down map is not sufficient for the sides of the structure or targets whether elevated or depressed and thus orbital flights at lower elevations are required .
I am testing with 3 lower orbits by returning to the DJIGO4 app

  1. Can DroneDeploy perform this without returning to DjiGO4 app
    EDIT- i believe not and have added “auto flight mode” but where this fails it seems clear i must manually go to DJIGo4, create an orbiral flight plan , at different elevations and calculate camera angle. its this relationship between elevation and camera angle that i don’t understand and welcome input about

2. Can DroneDeploy set suggested orbital altitudes , and camera angle
MY EDIT - I see not, and seek confirmation

**3. Can DroneDeploy set this entire mission as one autonomous project without needing to create additional projects, or possibly additional flight plans that may run separately but are set by DD as part of the same project
MY EDIT - i see only to the extent that it can address it in the one NADIR flight with 3D mode on but not as above. Please confirm

While i am hoping DD is my solution ( ignoring the model creation options) , and i have found my first drone deploy map test to be efficient, but without enough advanced capacity unless are there an alternatives to achieve my requirements which i should look at while DD goes this way. I am sure i wil be happy when i have found that i can easily capture the oblique shots in a DJIGO4 intellegient flight mode whether orbial or zipline

I have run a test with Autopilot with other problems but its much more complex to get skilled at so i have no opinion yet on this one other than it provides more advanced control, has a more complex but not very difficult learning curve, and doenst provide recommendations for altitudes, oblique shots , overlap or others that i can see

A multi-caputireflight as i am seeking when problems emerge with the first and only plan on DD seems essential ultimately and is certainly needed by me

thanks much in advance
input , chat welcome


#2

Sculpting 3D with a drone is part art and part science. I did this one by flying manual orbits with the DJI GO4 app at two altitudes shooting 85 overlapping oblique images looking down 45 – 60 degrees. I did not shoot any vertical grid images. Automated generic solutions in any form will not likely optimize the features of the buildings since they don’t know those features. Human interaction is usually still the best.


#3

Processing with DD, like this one, produced the best results.


#4

Hi thanks Gary
Great model , and i think its the schematic i studied by DroneDeploy with 4 passes
I think i am clear. DD can use the built in intelligence for the first mapping flight.
In my case as you suggest i must get in lower

i was hoping to do intellegent modes DJIGO POI or oribital flights
Thats ok for pass 1

However to get great quality down to the street level which can be quite narrow with obstacles i will need to do strightline plans or manual shots

I guess am trying to grasp h ow important distance and angles are for the additional pictures. When i get down to street level do i need to worry consider the distance from the side of the structure

  • It will be to the structure or is it useful to try to approximate the distance from the original flight model
  • it will lifkely be at about 20ft above ground where the original model was shot at 125 ft to accomodate the highest point
  • the camera angle will be quite different

1. Can i assume the cloud rendering will address this all acceptably and i do not need to worry about distance and angle?

2. Also, if i later upload a set of pictures of the side in the shadow, how does the software know to use the new pictures versus original or is it always going to try to upload the most recently uploaded pictures

3. is it correct that the DroneDeploy software cannot map an orbital flight and thus the lowest point for instance of a lighthouse must be the top of the lighthouse impacting the height, and resolution of the model negatively where it would be prefereably to have a lower orbital flight of the area? your example model would be a good example of this


#5

To sculpt a good 3D model it needs overlapping images preferably with ground features in them that it can align with. Horizontal shots don’t work as well.

It will use the images you upload. It may meld them.

No, it can map orbital lower than the top, just keep the camera pointing down - no sky in the images.