Yining Li

I am sorry that mmpose doesn't support 3D animal pose estimation. We don't have a plan to support it for now. Maybe this [paper](https://www.biorxiv.org/content/10.1101/476531v1.full.pdf) is related to your project.

Thanks for the feedback. We will look into the smoother and the demo script and try to figure out a better way to handle frames without detected objects. Please let...

请问具体指什么样的模块呢，可以举个例子么？

目前这方面没有太多支持，可以考虑看看 mmdetection 或者 mmclassification 里是否有你需要的模块。OpenMMLab 各 repo 之间的设计是比较一致的，应该可以轻松迁移其他 repo 里的模块到 mmpose 中。欢迎向 mmpose 提相关的 PR。

Hi, this issue will be closed for now. Please feel free to reopen it if there are further questions.

Please refer to our [multi-view demo](https://github.com/open-mmlab/mmpose/blob/master/demo/docs/3d_multiview_human_pose_demo.md). Please note that the camera parameter is required, and the model may not generalize very well on custom data which has different camera parameters...

Is it possible to feed video data frame-by-frame into the model? I guess the frames from different cameras can be temporally aligned with timestamps.

There are public tools to convert LabelMe data to COCO format, like https://github.com/fcakyon/labelme2coco. You can also write your own script to do the conversion. The description of COCO json format...

You can use `tools/test.py` for evaluation. Please refer to the [doc](https://mmpose.readthedocs.io/en/latest/get_started.html#inference-with-pre-trained-models) for details.

For now, we don't have such an evaluation method. You can follow the steps below to evaluate the model: 1. Organize your test data in a supported format (e.g. COCO)...