Cedric Menard

@RitchieQi this issue seems to arise from using the `package://` prefix, which is actually valid URDF description (see: http://wiki.ros.org/urdf/XML/link). Since urdfpy doesn't seem to support the package scheme, removing said...

IMO I don’t think it should be part of @fholger excellent tool kit. It is a hacky workaround. And at the end of the day, this kit still requires per-game...

Indeed, this is a quick fix and without it, it does not build on any platforms other than x86. Might be a good idea to investigate if some flags (such...

Unfortunately I haven't looked at this. Sorry.

Hi @shreyasashwath, You can take a look at the class definition in irislandmarks.py, especially at the predict and the forward methods. Training can be done as with any PyTorch models...

For face landmarks, I suggest looking at the [FaceMesh](https://github.com/thepowerfuldeez/facemesh.pytorch) repo. You can definitely use this model, but you'll have to modify it according to the output size you need. Batch...

I've made a quick change to python-fcl so I can use the `insertPointCloud` method of `octomap.OcTree`. See my fork (on a branch currently) + the source change: https://github.com/cedriclmenard/python-fcl-octomap/blob/ef7553eedc71fcd3f13a6717b7681b03f61f4418/src/fcl/fcl.pyx#L460

Once you create the `fcl.OcTree` object (without arguments), you can call `insertPointCloud` with a points list/array (n points x 3 dimension). The given OcTree will be compatible with all the...

> > Once you create the `fcl.OcTree` object (without arguments), you can call `insertPointCloud` with a points list/array (n points x 3 dimension). The given OcTree will be compatible with...