error about SIBR_viewer

[boxuLibrary]

hello, it's an excellent work. However, when I compile the render tools in the SIBR_viewer, I encounter a problem that I can not find the torch. Do you know how I can solve this problem? Thank you so much.

[tongji-rkr]

You need to download and configure libtorch(1.10) and choose the correct version according to your build mode (Debug or Release).

[tongji-rkr]

After downloading, compile according to the following process https://github.com/city-super/Scaffold-GS/tree/main/SIBR_viewers

[boxuLibrary]

Thanks, are there instructions for Linux? Because I am not familiar with windows.

[boxuLibrary]

When I run the original Gaussian Splatting, there is no similar problem. So I think the problem may be the Cmakelists?

[tongji-rkr]

Due to equipment reasons, we have not compiled it in linux before, but we will improve the Linux instructions in the near future.

[tongji-rkr]

Also, we figure out the cause of the problem. We'll work it out as soon as possible.

[RuiZ-18]

Hello! have you already figure out the linux version? I meet the same problem, and I've tried for a long time.

[bro-sin]

I met the same error and this could be solved appropriately using this compile command:

Torch_DIR=`python -c 'import torch;print(torch.utils.cmake_prefix_path)'` cmake -Bbuild . -DCMAKE_BUILD_TYPE=Release

if you have installed the pytorch package.

I encounter another problem that:

undefined reference to `cv::imread(std::string const&, int)'

I compiler this viewer using the following Dockerfile:

ARG CUDA_VERSION=11.7.1
ARG OS_VERSION=22.04
#编译viewers的时候需要使用CUDNN
ARG CUDNN_VERSION=8
FROM nvidia/cuda:${CUDA_VERSION}-cudnn${CUDNN_VERSION}-devel-ubuntu${OS_VERSION} as baseimage 

# Set environment variables.
## Set non-interactive to prevent asking for user inputs blocking image creation.
ENV DEBIAN_FRONTEND=noninteractive
## Set timezone as it is required by some packages.
ENV TZ=Asia/Shanghai
## CUDA Home, required to find CUDA in some packages.
ENV CUDA_HOME="/usr/local/cuda"

#换源
RUN sed -i "s/archive.ubuntu.com/mirrors.cqu.edu.cn/g" /etc/apt/sources.list \
    && sed -i "s/security.ubuntu.com/mirrors.cqu.edu.cn/g" /etc/apt/sources.list

#安装必要的软件
RUN apt-get update \
    && apt-get install -y --no-install-recommends \
    curl \
    git \
    build-essential \
    cmake \
    && rm -rf /var/lib/apt/lists/*

FROM baseimage as buildimage
#安装依赖
RUN apt-get update \
    && apt-get install -y --no-install-recommends \
    libglew-dev \
    libassimp-dev \
    libboost-all-dev \
    libgtk-3-dev \
    libopencv-dev \
    libglfw3-dev \
    libavdevice-dev \
    libavcodec-dev \
    libeigen3-dev \
    libxxf86vm-dev \
    libembree-dev \
    && rm -rf /var/lib/apt/lists/*

# 安装python以及python包
RUN apt-get update \
    && apt-get install -y --no-install-recommends \
    python3-dev \
    python3-pip \
    python-is-python3 \
    && rm -rf /var/lib/apt/lists/*
#换源
RUN pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
#安装torch
RUN pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1 \
    -f https://mirror.sjtu.edu.cn/pytorch-wheels/torch_stable.html

#添加源码
COPY . /sibr_core

#开始编译
WORKDIR /sibr_core

#设置环境变量,编译的时候会用到torch相关的库
# Variables used at build time.
ARG TORCH_CUDA_ARCH_LIST="3.5;5.0;6.0;6.1;7.0;7.5;8.0;8.6+PTX"

# cmake这一步会下载extlibs,与编译分开，便于进去调试
RUN mkdir build \
    && Torch_DIR=`python -c 'import torch;print(torch.utils.cmake_prefix_path)'` \
    cmake -Bbuild . -DCMAKE_BUILD_TYPE=Release
RUN  cmake --build build --target install

ENTRYPOINT [ "/sibr_core/install/bin/SIBR_gaussianViewer_app" ]

# how to build:
# docker build -t sibr_viewer:cudnn8-scaffold-gs -f docker/Dockerfile .

[tongji-rkr]

You can try to add -D_GLIBCXX_USE_CXX11_ABI=1 or -D_GLIBCXX_USE_CXX11_ABI=0 to your command.

[bro-sin]

You can try to add -D_GLIBCXX_USE_CXX11_ABI=1 or -D_GLIBCXX_USE_CXX11_ABI=0 to your command.

I tried both options, but neither worked. Additionally, when I added the option, the following warning appeared:

CMake Warning:
  Manually-specified variables were not used by the project:

    _GLIBCXX_USE_CXX11_ABI