Parsing-R-CNN
Parsing-R-CNN copied to clipboard
soeaver
Parsing R-CNN for Instance-Level Human Analysis
Thanks for your work. I tried to run `make.sh` in windows, namely, setup python extensions. However, I got some errors. Can you help me? P.S. My python version is 3.7....
when i use comand sudo python setup.py install --cuda_ext --cpp_ext error occured ： def raise_if_cuda_home_none(global_option: str) -> None:
I download MHP-v2 dataset from offical website, but it's not contructed in COCO format. Could you please upload your processed MHP-v2 dataset? Thanks a lot.
@soeaver @zhirihuixin thanks for sharing the code base , can we run the inference on cpu ? what all changes have to be made for the model to run on...
Hi @soeaver, Thank you for releasing the excellent codebase. I wanted to ask if you would be releasing weights for your state of art model with ResNext101 backend (for CIHP...
@soeaver @rs9899 Thanks for open sourcing the code base i have few queries 1. can we change the parsing branch to an attribute branch where age and gender are the...
As the title expressed, thank u a lot!
When I run the test code I got the error due to the mismatch of the weights size mismatch for RPN.anchor_generator.cell_anchors.0: copying a param with shape torch.Size([3, 4]) from checkpoint,...
when I run the CUDA_VISIBLE_DEVICES=0 python tools/train_net.py --cfg cfgs/DensePose_COCO/e2e_parsing_rcnn_R-50-FPN_s1x_ms.yaml commend,I only have one GPU,but the uv_ann_labels problems always cause the mistake as follows: Traceback (most recent call last): File "tools/train_net.py",...
