[feature] Add Dockerfile and instructions of installation by docker

Dockerfile

@@ -0,0 +1,31 @@
+ARG PYTORCH="1.8.0"
+ARG CUDA="11.1"
+ARG CUDNN="8"
+
+FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel
+
+ENV TORCH_CUDA_ARCH_LIST="6.0 6.1 7.0 8.0 8.6"
+ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
+ENV CMAKE_PREFIX_PATH="$(dirname $(which conda))/../"
+
+RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
+RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/7fa2af80.pub
+
+RUN apt-get update && apt-get install -y git ninja-build libglib2.0-0 libsm6 libxrender-dev libxext6 libgl1-mesa-glx \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN conda install ffmpeg -y
+RUN conda install -c fvcore -c iopath -c conda-forge fvcore iopath -y
+RUN conda install -c bottler nvidiacub -y
+RUN conda install pytorch3d -c pytorch3d -y
+RUN pip install "mmcv-full>=1.3.17,<=1.5.3" -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.8.0/index.html
+RUN pip install "mmdet<=2.25.1"
+WORKDIR /
+RUN git clone https://github.com/open-mmlab/mmhuman3d.git
+WORKDIR /mmhuman3d
+RUN pip install -v -e .
+
+RUN pip install numpy==1.23
+RUN pip install pillow==9.5
+RUN pip install scipy==1.9.1

docs/getting_started.md

@@ -5,8 +5,8 @@
   - [Data Preparation](#data-preparation)
   - [Body Model Preparation](#body-model-preparation)
   - [Inference / Demo](#inference--demo)
-    - [Offline Demo](#Offline-Demo)
-    - [Online Demo](#Online-Demo)
+    - [Offline Demo](#offline-demo)
+    - [Online Demo](#online-demo)
   - [Evaluation](#evaluation)
     - [Evaluate with a single GPU / multiple GPUs](#evaluate-with-a-single-gpu--multiple-gpus)
     - [Evaluate with slurm](#evaluate-with-slurm)
@@ -17,7 +17,7 @@
 
 ## Installation
 
-Please refer to [install.md](./install.md) for installation.
+Please refer to [install.md](./install.md) for installation or [install_by_docker.md](./install_by_docker.md) for installation through docker.
 
 ## Data Preparation
 

docs/install_by_docker.md

@@ -0,0 +1,29 @@
+# Installation by Docker
+
+After pull the image from [here](https://hub.docker.com/repository/docker/jacobbbbbb/mmhuman3d/general), simply run the following command:
+```
+docker run --gpus all --shm-size=64g -it -v {DATA_DIR}:/mmhuman3d/data mmhuman3d
+```
+
+We test this command on RTX 3090. Alternatively, you can build docker image based on your own GPU type by the following command with the provided Dockerfile:
+```
+docker image build -t mmhuman3d
+```
+
+Then run the command above. Note that inside the container, we donot provide data including body_models etc, please refer to [data_preparation](https://github.com/open-mmlab/mmhuman3d/blob/main/docs/preprocess_dataset.md) for preparation details.
+
+After data prepared, this [demo command](https://github.com/open-mmlab/mmhuman3d/blob/main/docs/getting_started.md) can be a quick check:
+```shell
+python demo/estimate_smpl.py \
+    configs/hmr/resnet50_hmr_pw3d.py \
+    data/checkpoints/resnet50_hmr_pw3d.pth \
+    --single_person_demo \
+    --det_config demo/mmdetection_cfg/faster_rcnn_r50_fpn_coco.py \
+    --det_checkpoint https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco/faster_rcnn_r50_fpn_1x_coco_20200130-047c8118.pth \
+    --input_path  demo/resources/single_person_demo.mp4 \
+    --show_path vis_results/single_person_demo.mp4 \
+    --output demo_result \
+    --smooth_type savgol \
+    --speed_up_type deciwatch \
+    --draw_bbox
+```