diff --git a/cv/point_cloud/Point-BERT/pytorch/README.md b/cv/point_cloud/Point-BERT/pytorch/README.md
index d37c9f9c85662d14107aed550c56ac57f9625cf8..4db8ccb5fad88d776209741953dbe9f714d36bb8 100644
--- a/cv/point_cloud/Point-BERT/pytorch/README.md
+++ b/cv/point_cloud/Point-BERT/pytorch/README.md
@@ -3,60 +3,74 @@ Point-BERT is a new paradigm for learning Transformers to generalize the concept
 
 ## Step 1: Installing packages
 
+> Warning: Now only support Ubuntu OS. If your OS is centOS, you may need to compile open3d from source.
+
 * system
 
 ```shell
-$ apt update
-$ apt install libgl1-mesa-glx
+apt update
+apt install libgl1-mesa-glx
 ```
 
 * python 
-```
-$ pip3 install argparse easydict h5py matplotlib numpy open3d==0.10 opencv-python pyyaml scipy tensorboardX timm==0.4.5  tqdm transforms3d termcolor scikit-learn==0.24.1 Ninja --default-timeout=1000
+
+```shell
+pip3 install argparse easydict h5py matplotlib numpy open3d==0.10 opencv-python pyyaml scipy tensorboardX timm==0.4.5  tqdm transforms3d termcolor scikit-learn==0.24.1 Ninja --default-timeout=1000
 ```
 
 * Chamfer Distance
-```
-$ cd /path/to/Point-BERT/pytorch
-$ bash install.sh
+
+```shell
+bash install.sh
 ```
 
 * PointNet++
-```
-$ cd ./Pointnet2_PyTorch
-$ pip3 install pointnet2_ops_lib/.
-$ cd -
+```shell
+cd ./Pointnet2_PyTorch
+pip3 install pointnet2_ops_lib/.
+cd -
 ```
 
 * GPU kNN
-```
-$ pip3 install --upgrade https://github.com/unlimblue/KNN_CUDA/releases/download/0.2/KNN_CUDA-0.2-py3-none-any.whl
+```shell
+pip3 install --upgrade https://github.com/unlimblue/KNN_CUDA/releases/download/0.2/KNN_CUDA-0.2-py3-none-any.whl
 ```
 
 
 ## Step 2: Preparing datasets
 
-Please reference [DATASET.md](./DATASET.md) to prepare `ShapeNet55` and `processed ModelNet`.
-
+Please refer to [DATASET.md](./DATASET.md) for preparing `ShapeNet55` and `processed ModelNet`.
+The dataset dircectory tree would be like:
+```shell
+data/
+├── ModelNet
+│   └── modelnet40_normal_resampled
+│       ├── modelnet40_test_8192pts_fps.dat
+│       └── modelnet40_train_8192pts_fps.dat
+├── ScanObjectNN_shape_names.txt
+├── ShapeNet55-34
+│   ├── ShapeNet-55
+│   │   ├── test.txt
+│   │   └── train.txt
+│   └── shapenet_pc
+└── shapenet_synset_dict.json
+```
 
 ## Step 3: Training
 
 * dVAE train
-```
-$ cd /path/to/Point-BERT
-$ bash scripts/train.sh 0 --config cfgs/ShapeNet55_models/dvae.yaml --exp_name dVAE
+```shell
+bash scripts/train.sh 0 --config cfgs/ShapeNet55_models/dvae.yaml --exp_name dVAE
 ```
 
 * Point-BERT pre-training
 
 When dVAE has finished training, you should be edit `cfgs/Mixup_models/Point-BERT.yaml`, and add the path of dvae_config-ckpt.
 
-```
-$ cd /path/to/Point-BERT
-$ bash ./scripts/dist_train_BERT.sh <NumGPUs> 12345 --config cfgs/Mixup_models/Point-BERT.yaml --exp_name pointBERT_pretrain --val_freq 2 
+```shell
+bash ./scripts/dist_train_BERT.sh <NumGPUs> 12345 --config cfgs/Mixup_models/Point-BERT.yaml --exp_name pointBERT_pretrain --val_freq 2 
 ```
 
-> Warning: You may need to compile open3d, when your os is centos.
 
 
 ## Reference