Yu Sun

Sorry for the late reply. @mkhoshle Yes, I have met this issue before. The problem is that some gpus don't have the meta_data input. The problem is caused by the...

During evaluation along with the training process, if would be fine to set `drop_last` to avoid this. During formal evaluation phrase, we can use only one gpu for evaluation to...

Hi, brother Tian, We also tried to drive the mixamo charactors, but this process need a retargeting algorithm to solve the joint rotation mapping. Unfortunately, we haven't solve this retargeting...

马普所的大佬们看不懂中文呀～

Sorry, never use the torch.utils.bottleneck. Don't know why.

HI, 1. Download link https://github.com/Arthur151/ROMP/releases/download/v1.1/trained_models.zip 2. Yes, it only contains backbone parameters. 3. If you already set hrnet_pretrain, then it could be properly loaded via https://github.com/Arthur151/ROMP/blob/d78880a1c56bd92df7e8f28d5d5aa19c6b7106f6/romp/lib/models/romp_model.py#L34 If you also set...

@simon-thu , 用HRNet的版本会比较好。我只试过L2 norm直接学，不过貌似并不是很合理，我主要是用来引导一下某些不好训的backbone的训练。

Yeah, Interesting feature. To achieve this, you only need to attach to a part index label to each vertex. Here is the thing you want https://github.com/Meshcapade/wiki/blob/main/assets/SMPL_body_segmentation/smpl/smpl_vert_segmentation.json

You can use the results to create a new dataset. To add a new dataset for training: 1. Follow the dataloader of an existing dataset, like [h36m](https://github.com/Arthur151/ROMP/blob/master/romp/lib/dataset/h36m.py), to write a...

Here are the relationship: imgpath: path to the input image. image: the loaded image via cv2.imread, in RGB order. kp2ds: can use the `pj2d_org` directly. track_ids: can be None, all...