pramishp

Get lens distortion characteristics

1

I have been using this app for quite a while and I am absolutely loving it. As per apple documentation "depth map representations are geometrically distorted to align with images...

About CPU inferance

2

thank you for open sourcing this amazing piece of work. I tried running the code on windows 10 without GPU with`use_cuda: false` and removing some cuda checks. But, it threw...

Kid body model

2

There is `kid_template_path params`. I tried providing `SMIL body model` to `kid_template_path` downloaded from https://www.iosb.fraunhofer.de/en/competences/image-exploitation/object-recognition/sensor-networks/motion-analysis.html#smil . But, this doesn't work. Can you please provide the link to download SMIL/SMIL-X model...

About Accuracy

1

What is the accuracy achieved with the current model? It would be very interesting to delve deep and motivating factor if you have found some correlation with astrological data.

RuntimeError: expected input[100, 3, 512, 32] to have 131 channels

8

I am trying to run the training on my Ubuntu 18.04.5 with python=3.7.10, torch=1.1.0. However, I am getting the following error when trying to run training with batchsize=100 ``` .........

Pre-trained Model

1

Is there plan to release pre-trained model ?

Pose params for SMPL model

2

With 50 and 53 pose params set to 5.6 and -5.6, the SMPL model gives pose as shown:  I am using [smplx](https://github.com/vchoutas/smplx) package with model_type='smpl'.

Thank you for open sourcing this great work ! Is there a plan to release the training code for uv-position control net ? Releasing this would help a lot with...