Jad Abou-Chakra
Jad Abou-Chakra
The scale is just transforming between real world units (the units your camera poses are in) to a number between 0 and 65535 (so it can be saved in a...
Are you sure your integer_depth_scale is correct? This is the number that is multiplied by your depth map to transform it into world units (the units your camera poses were...
If you're using an iOS device to capture this dataset, bear in mind that the intrinsics change with every frame due to the optical stabilization. The transforms.json allows you to...
Those are pretty much the only controls that I know of: The scale and the configs. To get a nice scale for your scene, I find it easier to just...
Hey @Tyler-D, as far as I remember - the hash encoding's parameter are simply what they outline in their paper. That is probably the best source on the subject. 
Just add "fl_x" "fl_y" "cx" "cy" to each frame in transforms.json. They will override the ones set above I believe.
Still having problems with this
This is amazing @JeanElsner. Another BIG thank you from me.
Hi Mateusz, The frames are updated through the python API at the moment, so you will have to use python particle_scripts/run.py to step through the animation instead of using instant-ngp...
Unfortunately, the save snapshot was not implemented - mainly because it was not needed for the original paper. The network however is being saved which explains the ghosting.