Yufei Xu

Hi@khanhha, Thanks for your interest in ViTPose. Hi, awesome developers from MMPose. Thanks for your fantastic work in MMPose. The framework is wonderful and helps greatly in our ViTPose and...

Hi. I got a similar 0.25G MACs for the T2T modules. I tried to calculate the MACs following the code and the [repo](https://github.com/google-research/electra/blob/master/flops_computation.py) as you suggested in [another issue](https://github.com/yitu-opensource/T2T-ViT/issues/10) for...

I tried the command `git reset --hard a651e2c24ecf97cbf367fd3f330df36760e1c597` to revert the repo to several days ago and this works for me.

After I downgrade the NDK version, the probelm solved for me

Hi, If the dataset is organized like ImageNet, i.e., ```text data |── train │── class1 │-- image1 │-- image2 │-- image3 |── val │── class1 │-- image1_v │-- image2_v │--...

Hi, As reported in the paper (Table 6), it can achieve 15 fps when stabilizing videos.

Hi! We simply convert the model to ONNX format, which has the TensorRT backend to acceleration. It should be noted that the latest PyTorch should be used for converting as...

Hi, Thanks for your interest. Please also make corresponding changes in find homography functions, where I assumed the input size is 640, 480, so the grid border for outliers is...

Hi, I remember that change several absolute values around [this line](https://github.com/Annbless/DUTCode/blob/eb623e2437a0074ffea46f8f39f39796f9aa084a/utils/ProjectionUtils.py#L131) is able to adapt the DUT code to smaller or larger resolution. To be specific, the value 39 comes...

Sorry for the late reply! The code and pre-trained models for all ViTAE variants are being organized. Hopefully, we will be able to release these models in late March or...