Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix links for cfg #264

Merged
merged 1 commit into from
Nov 9, 2020
Merged

fix links for cfg #264

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 6 additions & 6 deletions demo/2d_hand_demo.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ python demo/top_down_img_demo.py \

```shell
python demo/top_down_img_demo.py \
configs/top_down/resnet/onehand10k/res50_onehand10k_256x256.py \
configs/hand/resnet/onehand10k/res50_onehand10k_256x256.py \
https://download.openmmlab.com/mmpose/top_down/resnet/res50_onehand10k_256x256-e67998f6_20200813.pth \
--img-root tests/data/onehand10k/ --json-file tests/data/onehand10k/test_onehand10k.json \
--out-img-root vis_results
Expand Down Expand Up @@ -51,7 +51,7 @@ python demo/top_down_img_demo_with_mmdet.py \
```shell
python demo/top_down_img_demo_with_mmdet.py demo/mmdetection_cfg/cascade_rcnn_x101_64x4d_fpn_20e_onehand10k.py \
https://download.openmmlab.com/mmpose/mmdet_pretrained/cascade_rcnn_x101_64x4d_fpn_20e_onehand10k-dac19597_20201030.pth \
configs/top_down/resnet/onehand10k/res50_onehand10k_256x256.py \
configs/hand/resnet/onehand10k/res50_onehand10k_256x256.py \
https://download.openmmlab.com/mmpose/top_down/resnet/res50_onehand10k_256x256-e67998f6_20200813.pth \
--img-root tests/data/onehand10k/ \
--img 9.jpg \
Expand Down Expand Up @@ -83,7 +83,7 @@ Examples:
```shell
python demo/top_down_video_demo_with_mmdet.py demo/mmdetection_cfg/cascade_rcnn_x101_64x4d_fpn_20e_onehand10k.py \
https://download.openmmlab.com/mmpose/mmdet_pretrained/cascade_rcnn_x101_64x4d_fpn_20e_onehand10k-dac19597_20201030.pth \
configs/top_down/resnet/onehand10k/res50_onehand10k_256x256.py \
configs/hand/resnet/onehand10k/res50_onehand10k_256x256.py \
https://download.openmmlab.com/mmpose/top_down/resnet/res50_onehand10k_256x256-e67998f6_20200813.pth \
--video-path demo/demo_video.mp4 \
--out-video-root vis_results
Expand All @@ -92,6 +92,6 @@ python demo/top_down_video_demo_with_mmdet.py demo/mmdetection_cfg/cascade_rcnn_
### Speed Up Inference
Some tips to speed up MMPose inference:

For 2D hand pose estimation models, try to edit the config file.
1. set `flip_test=False` (line 56 in [res50](/configs/top_down/resnet/onehand10k/res50_onehand10k_256x256.py))
2. set `unbiased_decoding=False` (line 59 in [res50](/configs/top_down/resnet/onehand10k/res50_onehand10k_256x256.py))
For 2D hand pose estimation models, try to edit the config file. For example,
1. set `flip_test=False` in [hand-res50](/configs/hand/resnet/onehand10k/res50_onehand10k_256x256.py#L56).
2. set `unbiased_decoding=False` in [hand-res50](/configs/hand/resnet/onehand10k/res50_onehand10k_256x256.py#L59).
18 changes: 9 additions & 9 deletions demo/2d_human_pose_demo.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,12 +128,12 @@ python demo/bottom_up_video_demo.py \
### Speed Up Inference
Some tips to speed up MMPose inference:

For top-down models, try to edit the config file.
1. set `flip_test=False` (line 51 in [topdown-res50](/configs/top_down/resnet/coco/res50_coco_256x192.py))
2. set `unbiased_decoding=False` (line 54 in [topdown-res50](/configs/top_down/resnet/coco/res50_coco_256x192.py))

For bottom-up models, try to edit the config file.
1. set `flip_test=False` (line 80 in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py))
2. set `adjust=False` (line 78 in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py))
3. set `refine=False` (line 79 in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py))
4. use smaller input image size (line 39 in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py))
For top-down models, try to edit the config file. For example,
1. set `flip_test=False` in [topdown-res50](/configs/top_down/resnet/coco/res50_coco_256x192.py#L51).
2. set `unbiased_decoding=False` in [topdown-res50](/configs/top_down/resnet/coco/res50_coco_256x192.py#L54).

For bottom-up models, try to edit the config file. For example,
1. set `flip_test=False` in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py#L80).
2. set `adjust=False` in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py#L78).
3. set `refine=False` in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py#L79).
4. use smaller input image size in [bottomup-res50](/configs/bottom_up/resnet/coco/res50_coco_512x512.py#L39).