Issue with Running Video Pose Demo: “Please install mmdet to run the demo” and “ModuleNotFoundError: No module named ‘mmcv._ext’”

[JacklE0niden]

Hello!

I encountered issues when trying to run the video pose demo using mmdet. The error message is as follows:
"Please install mmdet to run the demo."

Additionally, when I try to run the following:

    from mmdet.apis import inference_detector, init_detector

I get the following error:

ModuleNotFoundError: No module named 'mmcv._ext'

Here is the command I used to run the demo:

python demo/top_down_video_demo_with_mmdet.py \
    demo/mmdetection_cfg/cascade_rcnn_x101_64x4d_fpn_1class.py \
    https://openmmlab.oss-cn-hangzhou.aliyuncs.com/mmpose/mmdet_pretrained/cascade_rcnn_x101_64x4d_fpn_20e_macaque-e45e36f5_20210409.pth \
    configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/res152_macaque_256x192.py \
    https://download.openmmlab.com/mmpose/animal/resnet/res152_macaque_256x192-c42abc02_20210407.pth \
    --video-path 'dogvideo/train_set/jump02.mp4' \
    --out-video-root '/dog_action/pose_estimation/output/ViTPose/trainset' \
    --bbox-thr 0.5 \
    --det-cat-id 17 \
    --kpt-thr 0.3 

Environment:
CUDA 12.4
torch==1.9.0
mmcv-full=1.5.0 mmdet=2.19.0

Could you please assist me with resolving this issue? It seems like there may be a mismatch or some dependency that is missing.