Yufei Xu

Hi, Thanks for your notice. In the ViTPose paper, the results are reported using the multi-task training setting. In our ViTPose++ paper (Table 15), the ViTPose results indicate the model...

Hi, We use zero padding around the patch embeding weight to get a 16x16 patch embedding from 14x14 patch embedding.

Thanks for your interest. I think the size you talk about is the size of the checkpoint, right? The default checkpoint saved contains not only the model's weights but also...

It seems there are no further questions. I will close this issue temporarily. If you have some further questions, please feel free to re-open it.

Thanks for your attention. Please refer to the paper for the settings in the speed test. With the advanced GPUs and PyTorch framework, ViTPose is faster than HRNet. Besides, the...

It is normal to have these warnings of mismatched keys as we do not use the cls token in the backbone model. It is wired for the 2nd problem. It...

It seems it is more related to the git configuration and does not affect the training. Please just ignore it. I will close this issue temporarily. If you have any...

The different configurations of our models can be found in the [folder](https://github.com/ViTAE-Transformer/ViTPose/tree/main/configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco). In each config file, there is a dictionary named `model`, which contains two sub-dictionaries named `backbone` and `keypoint...

It seems there are no further questions. I will close this issue temporarily. If you have any further questions, please feel free to re-open it.

Hi, you can try the web demo here [link](https://huggingface.co/spaces/hysts/ViTPose_video). As demonstrated in the error msg, it seems that the error is caused that the installed package is ```mmcv``` not ```mmcv-full```....