Yufei Xu
Yufei Xu
Hi, Thanks for your interest! The code will definitely be made public in full in the near future, but I'm not sure exactly when due to the many events that...
Sorry for the late reply. I am afraid that the CNN-based keypoint detection model cannot extract enough keypoints in such areas without texture and edges. I visualized the keypoints extracted...
Not exactly. Please refer to Figures 3 and 5 in the paper. The distortion metric only describes the global artifacts as the frame ratio changes, but it does not measure...
Hi, Please refer to the [Link](https://github.com/ViTAE-Transformer/). We have released the code for ViTAE/ViTAEv2/VSA/remote sensing/matting/text recognition/pose estimation/ViTDet/ to facilitate the usage.
Hi, This seems to be due to a memory issue with Colab ....... I am not sure if there are any changes in the usage limits of Colab. I will...
It seems that temporarily mmpose needs at least one GPU for test...
Thanks for your comments. As demonstrated in Table 6, the performance gains brought by crowdpose is relatively small compared with the performance gains brought by AIC. We suspect that the...
Hi, We updated the training results for ViTPose-B, ViTPose-L, and ViTPose-H without images from the CrowdPose dataset. Due to the reduction of duplicate images and in the number of training...
It seems there are no further questions. I will close this issue temporarily. If you have any more questions, please feel free to re-open it.
Hi. The checkpoints can be found in [link](https://1drv.ms/u/s!AimBgYV7JjTlgcdFowH8IVITrl8jhw?e=8nodAk).