diff --git a/configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-27frm-supv_8xb128-120e_h36m.py b/configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-27frm-supv_8xb128-160e_h36m.py similarity index 100% rename from configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-27frm-supv_8xb128-120e_h36m.py rename to configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-27frm-supv_8xb128-160e_h36m.py diff --git a/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.md b/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.md index 815b91827b..8f7c8fba5b 100644 --- a/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.md +++ b/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.md @@ -41,7 +41,7 @@ Testing results on Human3.6M dataset with ground truth 2D detections, supervised | Arch | Receptive Field | MPJPE | P-MPJPE | ckpt | log | | :--------------------------------------------------------- | :-------------: | :---: | :-----: | :--------------------------------------------------------: | :-------------------------------------------------------: | -| [VideoPose3D-supervised-27frm](/configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-27frm-supv_8xb128-120e_h36m.py) | 27 | 40.1 | 30.1 | [ckpt](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_supervised-fe8fbba9_20210527.pth) | [log](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_supervised_20210527.log.json) | +| [VideoPose3D-supervised-27frm](/configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-27frm-supv_8xb128-160e_h36m.py) | 27 | 40.1 | 30.1 | [ckpt](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_supervised-fe8fbba9_20210527.pth) | [log](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_supervised_20210527.log.json) | | [VideoPose3D-supervised-81frm](/configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-81frm-supv_8xb128-160e_h36m.py) | 81 | 39.1 | 29.3 | [ckpt](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_81frames_fullconv_supervised-1f2d1104_20210527.pth) | [log](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_81frames_fullconv_supervised_20210527.log.json) | | [VideoPose3D-supervised-243frm](/configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-243frm-supv_8xb128-160e_h36m.py) | 243 | 37.6 | 28.3 | [ckpt](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_243frames_fullconv_supervised-880bea25_20210527.pth) | [log](https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_243frames_fullconv_supervised_20210527.log.json) | diff --git a/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.yml b/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.yml index 6b9d92c115..2ae0513699 100644 --- a/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.yml +++ b/configs/body_3d_keypoint/pose_lift/h36m/videopose3d_h36m.yml @@ -6,13 +6,13 @@ Collections: URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Pavllo_3D_Human_Pose_Estimation_in_Video_With_Temporal_Convolutions_and_CVPR_2019_paper.html README: https://github.com/open-mmlab/mmpose/blob/main/docs/en/papers/algorithms/videopose3d.md Models: -- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-243frm-supv_8xb128-80e_h36m.py +- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-27frm-supv_8xb128-160e_h36m.py In Collection: VideoPose3D Metadata: Architecture: &id001 - VideoPose3D Training Data: Human3.6M - Name: pose-lift_videopose3d-243frm-supv_8xb128-80e_h36m + Name: pose-lift_videopose3d-27frm-supv_8xb128-160e_h36m Results: - Dataset: Human3.6M Metrics: @@ -20,7 +20,7 @@ Models: P-MPJPE: 30.1 Task: Body 3D Keypoint Weights: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_supervised-fe8fbba9_20210527.pth -- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-81frm-supv_8xb128-80e_h36m.py +- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-81frm-supv_8xb128-160e_h36m.py In Collection: VideoPose3D Metadata: Architecture: *id001 @@ -33,7 +33,7 @@ Models: P-MPJPE: 29.2 Task: Body 3D Keypoint Weights: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_81frames_fullconv_supervised-1f2d1104_20210527.pth -- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-243frm-supv_8xb128-80e_h36m.py +- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-243frm-supv_8xb128-160e_h36mpy In Collection: VideoPose3D Metadata: Architecture: *id001 @@ -46,7 +46,7 @@ Models: P-MPJPE: 28.3 Task: Body 3D Keypoint Weights: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_243frames_fullconv_supervised-880bea25_20210527.pth -- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-1frm-supv-cpn-ft_8xb128-80e_h36m.py +- Config: configs/body_3d_keypoint/pose_lift/h36m/pose-lift_videopose3d-1frm-supv-cpn-ft_8xb128-160e_h36m.py In Collection: VideoPose3D Metadata: Architecture: *id001