Marek

Hi, If you do not need the functionality provided by the functions that are not present, you can return some sort of an empty input. Thanks, Marek

Hi, Yes the data can be saved to .ply format. During recording the .bin file is treated as a temporary buffer by the client. Once the recording is finished, the...

Hi Chien, That's a good point. While in the current version the skeleton data is not saved during sequence recording, this might be confusing to someone who wants to add...

Hi! Do you start both the client and the server? When you start the client, is there any message on the status bar saying why the Kinect did not start?...

Hi, As I mentioned before you need to start the **client** not the player. The player is for playing prerecorded sequences. It does not run the Kinect, that's why the...

Hi! I know that this one might be obvious, but have you look around the room (including behind you) to see if the reconstructed point cloud is there? This was...

I haven't heard of anyone trying. The depth data can be obtained via the [research mode](https://docs.microsoft.com/en-us/windows/mixed-reality/develop/platform-capabilities-and-apis/research-mode).

Hi, In the paper DAN is tested on a two datasets 300-W and Menpo. The Menpo dataset does not have face detection bounding boxes, so for this dateset I decided...

The stages were separated to allow the first stage to learn fully before the second stage is added. Thanks to this approach the second stage doesn't start lerning before the...

Hi, As for 1 in your custom layer you should - pass the variable named input as the first parameter to separable_conv2d (instead of x_sym) - as for the dfilter_sym...