Yining Li

May need a rebase to fix the lint issue in CI.

We may need to rethink the name "RootDisplacement", which I am afraid is not straightforward for some users. What about SPR (Structured Pose Representation) as it's termed in the SPM...

Hi, you may need to implement a new node for TensorTR model inference, just like [`TopdownPoseEstimatorNode` ](https://github.com/open-mmlab/mmpose/blob/1.x/mmpose/apis/webcam/nodes/model_nodes/pose_estimator_node.py#L22)for pytorch model inference. BTW, it would be great if you would like to...

Among these models, ssdlite and yolov3 should have faster inference speeds. You can also have a look at the MMDetection model zoo where there are more available detection models: https://mmdetection.readthedocs.io/en/3.x/model_zoo.html...

If you run `webcam_demo.py` with the [default config](https://github.com/open-mmlab/mmpose/blob/1.x/demo/webcam_cfg/pose_estimation.py) which has the MonitorNode, the diagnostic information can be enabled by pressing the key "m" and will be shown on the image.

Hi, thanks for the questions and feedback. 1. We suggest adding `CUDA_VISIBLE_DEVICES=2` at the beginning of the command line to specify the GPU index. 2. It seems that these are...

Hi, the warning has been fixed in the latest dev-1.x code. Is there any progress on the GPU selection issue?

There should be no problem accessing the variable `its` outside the for-loop as long as L76 has been reached. Did you encounter any errors or get incorrect results here?

Thanks for the feedback. We will check this issue soon. Are these `metadata.xml` and data files obtained from the Human3.6M official site?

Videopose3D is a pose-lifting algorithm that lifts 2D keypoint coordinates to 3D without using image data. Details can be found in the paper: https://arxiv.org/abs/1811.11742