Merge dev-0.21 into master (#1062)

* map location to cpu when load checkpoint (#1007)

* [Enhancement] Support minus output feature index in mobilenet_v3 (#1005)

* fix typo in mobilenet_v3

* fix typo in mobilenet_v3

* use -1 to indicate output tensors from final stage

* support negative out_indices

* [Enhancement] inference speed and flops tools. (#986)

* add the function to test the dummy forward speed of models.

* add tools to test the flops and inference speed of multiple models.

* [Fix] Update pose tracking demo to be compatible with latest mmtrakcing (#1014)

* update mmtracking demo

* support both track_bboxes and track_results

* add docstring

* [Fix] fix skeleton_info of coco wholebody dataset (#1010)

* fix wholebody base dataset

* fix lint

* fix lint

Co-authored-by: ly015 <[email protected]>

* [Feature] Add ViPNAS models for wholebody keypoint detection (#1009)

* add configs

* add dark configs

* add checkpoint and readme

* update webcam demo

* fix model path in webcam demo

* fix unittest

* [Fix] Fix bbox label visualization (#1020)

* update model metafiles (#1001)

* update hourglass ae .md (#1027)

* [Feature] Add ViPNAS mbv3 (#1025)

* add vipnas mbv3

* test other variants

* submission for mmpose

* add unittest

* add readme

* update .yml

* fix lint

* rebase

* fix pytest

Co-authored-by: jin-s13 <[email protected]>

* [Enhancement] Set a random seed when the user does not set a seed (#1030)

* fix randseed

* fix lint

* fix import

* fix isort

* update yapf hook

* revert yapf version

* add cfg file for flops and speed test,  change the bulid_posenet to init_pose_model and fix some typo in cfg (#1028)

* [Enhancement] Add more functions for speed test tool (#1034)

* add batch size and device args in speed test script, and remove MMDataParallel warper

* add vipnas_mbv3 model

* fix dead link (#1038)

* Skip CI when some specific files were changed (#1041)

* update sigmas (#1040)

* add more configs, ckpts and logs for HRNet on PoseTrack18 (#1035)

* [Feature] Add PoseWarper dataset (#1006)

* add PoseWarper dataset and base class

* modify pipelines related to video

* add unittest for PoseWarper dataset

* add unittest for evaluation function in posetrack18-realted dataset, and add some annotations json files

* fix typo

* fix unittest CI failure

* fix typo

* add PoseWarper dataset and base class

* modify pipelines related to video

* add unittest for PoseWarper dataset

* add unittest for evaluation function in posetrack18-realted dataset, and add some annotations json files

* fix typo

* fix unittest CI failure

* fix typo

* modify some methods in the base class to improve code coverage rate

* recover some mistakenly-deleted notes

* remove test_dataset_info part for the new TopDownPoseTrack18VideoDataset class

* cancel uncompleted previous runs (#1053)

* [Doc] Add inference speed results (#1044)

* add docs related to inference speed results

* add corresponding Chinese docs and fix some typos

* add Chinese docs in readthedocs

* remove the massive table in readme

* minor modification to wording

Co-authored-by: ly015 <[email protected]>

* [Feature] Add PoseWarper detector model (#932)

* Add top down video detector module

* Add PoseWarper neck

* add function _freeze_stages

* fix typo

* modify PoseWarper detector and PoseWarperNeck

* fix typo

* modify posewarper detector and neck

* Delete top_down_video.py

change the base class of `PoseWarper` detector from `TopDownVideo` to `TopDown`

* fix spell typo

* modify detector and neck

* add unittest for detector and neck

* modify unittest for posewarper forward

* Add top down video detector module

* Add PoseWarper neck

* add function _freeze_stages

* fix typo

* modify PoseWarper detector and PoseWarperNeck

* fix typo

* modify posewarper detector and neck

* Delete top_down_video.py

change the base class of `PoseWarper` detector from `TopDownVideo` to `TopDown`

* fix spell typo

* modify detector and neck

* add unittest for detector and neck

* modify unittest for posewarper forward

* modify dependency on mmcv version in posewarper neck

* reduce memory cost in test

* modify flops tool for more flexible input format

* Add top down video detector module

* Add PoseWarper neck

* add function _freeze_stages

* fix typo

* modify PoseWarper detector and PoseWarperNeck

* fix typo

* modify posewarper detector and neck

* Delete top_down_video.py

change the base class of `PoseWarper` detector from `TopDownVideo` to `TopDown`

* fix spell typo

* modify detector and neck

* add unittest for detector and neck

* modify unittest for posewarper forward

* Add PoseWarper neck

* modify PoseWarper detector and PoseWarperNeck

* modify posewarper detector and neck

* Delete top_down_video.py

change the base class of `PoseWarper` detector from `TopDownVideo` to `TopDown`

* fix spell typo

* modify detector and neck

* add unittest for detector and neck

* modify unittest for posewarper forward

* modify dependency on mmcv version in posewarper neck

* reduce memory cost in test

* modify flops tool for more flexible input format

* modify the posewarper detector description

* modify some arguments and related fields

* modify default values for some args

* fix readthedoc bulid typo

* fix ignore path (#1059)

* [Doc]  Add related docs for PoseWarper (#1036)

* add related docs for PoseWarper

* add related readme docs for posewarper

* modify related args in posewarper stage2 config

* modify posewarper stage2 config path

* add description about val_boxes path for data preparation (#1060)

* bump version to v0.21.0 (#1061)

* [Feature] Add ViPNAS_Mbv3 wholebody model (#1055)

* add vipnas mbv3 coco_wholebody

* add vipnas mbv3 coco_wholebody md&yml

* fix lint

Co-authored-by: ly015 <[email protected]>

Co-authored-by: Lumin <[email protected]>
Co-authored-by: zengwang430521 <[email protected]>
Co-authored-by: Jas <[email protected]>
Co-authored-by: jin-s13 <[email protected]>
Co-authored-by: Qikai Li <[email protected]>
Co-authored-by: QwQ2000 <[email protected]>

Author: 6 people

.dev_scripts/benchmark/benchmark_cfg.yaml

@@ -42,6 +42,7 @@ model_list:
     ## SimpleBaseline3D + H36M
     - config: configs/body/3d_kpt_sview_rgb_img/pose_lift/h36m/simplebaseline3d_h36m.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/simple_baseline/simple3Dbaseline_h36m-f0ad73a4_20210419.pth
+      input_shape: 34, 1
     # UDP
     ## AE + HRNet + UDP + COCO
     - config: configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/hrnet_w32_coco_512x512_udp.py
@@ -50,6 +51,7 @@ model_list:
     ## VideoPose3D + H36M
     - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/h36m/videopose3d_h36m_27frames_fullconv_supervised.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_supervised-fe8fbba9_20210527.pth
+      input_shape: 34, 27
     # MSPN
     ## MSPN + COCO
     - config: configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/mspn50_coco_256x192.py
@@ -114,6 +116,7 @@ model_list:
     ## SimpleBaseline3D + MPI-INF-3DHP
     - config: configs/body/3d_kpt_sview_rgb_img/pose_lift/mpi_inf_3dhp/simplebaseline3d_mpi-inf-3dhp.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/simplebaseline3d/simplebaseline3d_mpi-inf-3dhp-b75546f6_20210603.pth
+      input_shape: 34, 1
     # UDP
     ## AE + HigherHRNet + UDP + COCO
     - config: configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/higherhrnet_w32_coco_512x512_udp.py
@@ -135,6 +138,7 @@ model_list:
     ## VideoPose3D + MPI-INF-3DHP
     - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/mpi_inf_3dhp/videopose3d_mpi-inf-3dhp_1frame_fullconv_supervised_gt.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_mpi-inf-3dhp_1frame_fullconv_supervised_gt-d6ed21ef_20210603.pth
+      input_shape: 34, 1
     # LiteHRNet
     ## LiteHRNet + MPII
     - config: configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/mpii/litehrnet_18_mpii_256x256.py
@@ -190,7 +194,7 @@ model_list:
       checkpoint: https://download.openmmlab.com/mmpose/face/darkpose/hrnetv2_w18_wflw_256x256_dark-3f8e0c2c_20210125.pth
     - config: configs/hand/2d_kpt_sview_rgb_img/topdown_heatmap/onehand10k/hrnetv2_w18_onehand10k_256x256_dark.py
       checkpoint: https://download.openmmlab.com/mmpose/hand/dark/hrnetv2_w18_onehand10k_256x256_dark-a2f80c64_20210330.pth
-    - config: configs/hand/2d_kpt_sview_rgb_img/topdown_heatmap/panoptic2d/hrnetv2_w18_panoptic_256x256_dark.py
+    - config: configs/hand/2d_kpt_sview_rgb_img/topdown_heatmap/panoptic2d/hrnetv2_w18_panoptic2d_256x256_dark.py
       checkpoint: https://download.openmmlab.com/mmpose/hand/dark/hrnetv2_w18_panoptic_256x256_dark-1f1e4b74_20210330.pth
     - config: configs/hand/2d_kpt_sview_rgb_img/topdown_heatmap/rhd2d/hrnetv2_w18_rhd2d_256x256_dark.py
       checkpoint: https://download.openmmlab.com/mmpose/hand/dark/hrnetv2_w18_rhd2d_256x256_dark-4df3a347_20210330.pth
@@ -202,7 +206,7 @@ model_list:
     - config: configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/higherhrnet_w48_coco_512x512_udp.py
       checkpoint: https://download.openmmlab.com/mmpose/bottom_up/higher_hrnet48_coco_512x512_udp-7cad61ef_20210222.pth
     ## HRNetV2 + UDP + Panoptic
-    - config: configs/hand/2d_kpt_sview_rgb_img/topdown_heatmap/panoptic2d/hrnetv2_w18_panoptic_256x256_udp.py
+    - config: configs/hand/2d_kpt_sview_rgb_img/topdown_heatmap/panoptic2d/hrnetv2_w18_panoptic2d_256x256_udp.py
       checkpoint: https://download.openmmlab.com/mmpose/hand/udp/hrnetv2_w18_panoptic_256x256_udp-f9e15948_20210330.pth
     # SCNet
     # SCNet + COCO
@@ -219,17 +223,23 @@ model_list:
     ## VideoPose3D + H36M
     - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/h36m/videopose3d_h36m_81frames_fullconv_supervised.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_81frames_fullconv_supervised-1f2d1104_20210527.pth
+      input_shape: 34, 81
     - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/h36m/videopose3d_h36m_243frames_fullconv_supervised.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_243frames_fullconv_supervised_cpn_ft-88f5abbb_20210527.pth
+      input_shape: 34, 243
     - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/h36m/videopose3d_h36m_1frame_fullconv_supervised_cpn_ft.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_1frame_fullconv_supervised_cpn_ft-5c3afaed_20210527.pth
-    - config: cconfigs/body3d/videopose/h36m/videopose3d_h36m_243frames_fullconv_supervised_cpn_ft.py
+      input_shape: 34, 1
+    - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/h36m/videopose3d_h36m_243frames_fullconv_supervised_cpn_ft.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_243frames_fullconv_supervised_cpn_ft-88f5abbb_20210527.pth
+      input_shape: 34, 243
     - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/h36m/videopose3d_h36m_27frames_fullconv_semi-supervised.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_semi-supervised-54aef83b_20210527.pth
+      input_shape: 34, 27
     - config: configs/body/3d_kpt_sview_rgb_vid/video_pose_lift/h36m/videopose3d_h36m_27frames_fullconv_semi-supervised_cpn_ft.py
       checkpoint: https://download.openmmlab.com/mmpose/body3d/videopose/videopose_h36m_27frames_fullconv_semi-supervised_cpn_ft-71be9cde_20210527.pth
-    # Associative Embedding
+      input_shape: 34, 27
+  # Associative Embedding
     ## TODO: Other AE models
     ## AE + ResNet + COCO
     - config: configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/res50_coco_512x512.py