-
Hi, thank you for releasing code. It's a nice work. Can you provide the pose landmarks and heatmaps of Market1501?
-
Hello, I would like to consult the relevant details of gssl training. May I ask whether a net is initialized for each task when using gssl training in the paper, and is the latest training net predict…
-
ping @NicoKiaru
The aim of this proposal is to provide means of defining pairwise registrations of sources.
In case of grouping sets of sources into a common viewing scenario (multi-modal experi…
-
Description: I am unable to view the phantom model in fCal, which prevents me from performing phantom registration to obtain the PhantomToReference transformation matrix. This limitation is difficulti…
-
Add the ability to load offline models, so that users can still run their projects with spotty Internet / from places of the world where access to online models is restricted, i.e. China.
See prior…
-
### Have I written custom code (as opposed to using a stock example script provided in MediaPipe)
No
### OS Platform and Distribution
Linux Ubuntu 20.04
### MediaPipe version
0.10.10
…
-
Hello, how much data do you used for training? I marked 87 pictures, and the training does not converge for 100epoch
-
when i run the command
`roslaunch d2vins quadcam.launch enable_pgo:=true enable_loop:=true
`
the following error appeares
 that may be used as building blocks.
Re-us…