未验证 提交 436627d1 编写于 作者: W wuyefeilin 提交者: GitHub

update README.md and nets

update README.md and nets
...@@ -44,10 +44,10 @@ python data/download_data.py ...@@ -44,10 +44,10 @@ python data/download_data.py
## 快速体验视频流人像分割 ## 快速体验视频流人像分割
```bash ```bash
# 通过电脑摄像头进行实时分割处理 # 通过电脑摄像头进行实时分割处理
python video_infer.py --model_dir pretrained_weights/humanseg_lite_export python video_infer.py --model_dir pretrained_weights/humanseg_lite_inference
# 对人像视频进行分割处理 # 对人像视频进行分割处理
python video_infer.py --model_dir pretrained_weights/humanseg_lite_export --video_path data/video_test.mp4 python video_infer.py --model_dir pretrained_weights/humanseg_lite_inference --video_path data/video_test.mp4
``` ```
**NOTE**: **NOTE**:
...@@ -62,7 +62,7 @@ python train.py --model_type HumanSegMobile \ ...@@ -62,7 +62,7 @@ python train.py --model_type HumanSegMobile \
--data_dir data/mini_supervisely \ --data_dir data/mini_supervisely \
--train_list data/mini_supervisely/train.txt \ --train_list data/mini_supervisely/train.txt \
--val_list data/mini_supervisely/val.txt \ --val_list data/mini_supervisely/val.txt \
--pretrained_weights pretrained_weights/humanseg_mobile \ --pretrained_weights pretrained_weights/humanseg_mobile_ckpt \
--batch_size 8 \ --batch_size 8 \
--learning_rate 0.001 \ --learning_rate 0.001 \
--num_epochs 10 \ --num_epochs 10 \
......
from .backbone import mobilenet_v2 from .backbone import mobilenet_v2
from .backbone import xception from .backbone import xception
from .unet import UNet
from .deeplabv3p import DeepLabv3p from .deeplabv3p import DeepLabv3p
from .shufflenet_slim import ShuffleSeg from .shufflenet_slim import ShuffleSeg
from .hrnet import HRNet from .hrnet import HRNet
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册