Leaderboard

So it is able to -- in real time -- work through the following steps: Figure out the 3D position of every pixel in the terrain that is has just scanned Form the 3D topology from those pixel positions Identify the gaps in that topology (or calculate the discontinuity in the surface mesh) Reverse calculate where the best camera viewing angle would be to fill those gaps Translate that viewing angle back into a real world GPS coordinate position, and finally Direct the drone to that position and angle. Once completed, the algorithm would have to recalculate steps 1 to 3 again to insure that gap was successfully closed before moving to the next gap. Well....color me amazed. I would imagine that all the drone pilot has to do is point the drone toward the main target, define a preliminary flight path around that initial target position and then make sure there is enough memory storage and power for the drone to complete its task. At this rate, pretty soon the only thing the VFX artist or team will have to worry about is whether or not there are enough Keurig cups in the breakroom. Dave