yfeng95/delta

Learning Disentangled Avatars with Hybrid 3D Representations. (Face, Body, Hair and Clothing)

delta's Issues

error in building environment

When I install pytorch extension of tiny-cuda-nn,I encounter:
FAILED: /home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/bindings/torch/build/temp.linux-x86_64-cpython-39/tinycudann/bindings.o
c++ -MMD -MF /home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/bindings/torch/build/temp.linux-x86_64-cpython-39/tinycudann/bindings.o.d -pthread -B /media/data4/yejr/conda_env/delta/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /media/data4/yejr/conda_env/delta/include -I/media/data4/yejr/conda_env/delta/include -fPIC -O2 -isystem /media/data4/yejr/conda_env/delta/include -fPIC -I/home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/include -I/home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/dependencies -I/home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/dependencies/cutlass/include -I/home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/dependencies/cutlass/tools/util/include -I/home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/dependencies/fmt/include -I/media/data4/yejr/conda_env/delta/lib/python3.9/site-packages/torch/include -I/media/data4/yejr/conda_env/delta/lib/python3.9/site-packages/torch/include/torch/csrc/api/include -I/media/data4/yejr/conda_env/delta/lib/python3.9/site-packages/torch/include/TH -I/media/data4/yejr/conda_env/delta/lib/python3.9/site-packages/torch/include/THC -I/usr/local/cuda-11.5/include -I/media/data4/yejr/conda_env/delta/include/python3.9 -c -c /home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/bindings/torch/tinycudann/bindings.cpp -o /home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/bindings/torch/build/temp.linux-x86_64-cpython-39/tinycudann/bindings.o -std=c++17 -DTCNN_PARAMS_UNALIGNED -DTCNN_MIN_GPU_ARCH=86 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=_86_C -D_GLIBCXX_USE_CXX11_ABI=0
In file included from /home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/bindings/torch/tinycudann/bindings.cpp:44:0:
/home/yejr/Digital_Avater/DELTA-main/tiny-cuda-nn-master/dependencies/json/json.hpp:3954:14: fatal error: filesystem: No such file or directory
#include
^~~~~~~~~~~~
And my gcc version is 7.5.0,how should I deal with it?

The dataloader is yet to be released -- correct?

hi
great work !! This seems relevant to me. I suppose the dataloader is yet to be released? Or am I missing something.

from .dataset import build_dataset
ModuleNotFoundError: No module named 'lib.dataset'

thanks.

牛逼，我也点个赞

Transfering facial expressions

Thanks for the great work; how would you suggest performing expression transfer? do you have any insight on performance on this specific task compared to other models?

Is it possible to optimise the SMPL with Perspective Camera instead of Orthographic?

牛逼，点个赞。

How long does it take? 3090 can run?

How long does it take? 3090 can run?

Missing files

Hi authors,

Thanks for the great work! I have a few questions:

1. I noticed the repo only has configs and data for face training. What about full body/body training?
2. If we need to train body with SCARF, will NGP be added to SCARF?

How can I get fully texture body with hairstyle?

Hi, thank you for sharing your work. I have three questions about your model.

The first question is how can I run the model to get both body with fully textured and head with hairstyle.

The second question is whether it is possible to get a 3D obj format with body and head ?

The final question is can we test with our own datasets?

Thanks.