一、安装mmcv-full
GitHub - open-mmlab/mmcv: OpenMMLab Computer Vision Foundation
不支持torch低的版本。
本机环境cuda10.1，python3.7，torch1.7.1，torchvision0.8.2 对应下载链接：
pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu101/torch1.7.1/index.html
pip install mmpycocotools
pip install moviepy terminaltables seaborn decord -i https://pypi.douban.com/simple
自行安装opencv：https://blog.csdn.net/qq_34717531/article/details/107763872?spm=1001.2014.3001.5502二、安装 mmaction2
下载mmaction2 安装包：YFwinstony/JN-OpenLib-mmaction2
cd JN-OpenLib-mmaction2-main
python setup.py develop
?
注：会提示版本不符合
修改?? __init__.py?? ，
mmcv_maximum_version = '1.5.0'即可 ，我的mvcc-full版本是1.5.0
import mmcv
from .version import __version__, short_version
def digit_version(version_str):
digit_version = []
for x in version_str.split('.'):
if x.isdigit():
digit_version.append(int(x))
elif x.find('rc') != -1:
patch_version = x.split('rc')
digit_version.append(int(patch_version[0]) - 1)
digit_version.append(int(patch_version[1]))
return digit_version
mmcv_minimum_version = '1.2.4'
mmcv_maximum_version = '1.5.0'
mmcv_version = digit_version(mmcv.__version__)
assert (mmcv_version >= digit_version(mmcv_minimum_version)
and mmcv_version <= digit_version(mmcv_maximum_version)), \
f'MMCV=={mmcv.__version__} is used but incompatible. ' \
f'Please install mmcv>={mmcv_minimum_version}, <={mmcv_maximum_version}.'
__all__ = ['__version__', 'short_version']三、测试
准备一个视频 4.mp4，执行：
python demo/demo_spatiotemporal_det.py --video demo/4.mp4 --config configs/detection/ava/slowonly_omnisource_pretrained_r101_8x8x1_20e_ava_rgb.py --checkpoint https://download.openmmlab.com/mmaction/detection/ava/slowonly_omnisource_pretrained_r101_8x8x1_20e_ava_rgb/slowonly_omnisource_pretrained_r101_8x8x1_20e_ava_rgb_20201217-16378594.pth --det-config demo/faster_rcnn_r50_fpn_2x_coco.py --det-checkpoint http://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_2x_coco/faster_rcnn_r50_fpn_2x_coco_bbox_mAP-0.384_20200504_210434-a5d8aa15.pth --det-score-thr 0.9 --action-score-thr 0.5 --label-map demo/label_map_ava.txt --predict-stepsize 8 --output-stepsize 4 --output-fps 6 --out-filename demo/out_4.mp4结果：
?
