liheyuan

icon indicating a website link https://lhyfst.github.io/ icon indicating an email [email protected]

icon location Singapore icon location Simple is better than complex.

Results 36 comments of liheyuan

Facial Landmarks

@Elsaam2y Have you done the modification of the training code?

Prepare dataset

@littlePrince126 Can you reproduce the reported result on NoW using your generated data? Additionally, could you please share vggface2_val_list_max_normal_100_ring_5_1_serial.npy with me? Even a demo is helpful! I just want to...

The dataset reads multiple images

same question

Failure cases and cropping

Good observation! I found them, too.

train error--num_samples=0

same problem!

Meta Data Files of UCF 101

https://github.com/kenshohara/3D-ResNets-PyTorch#ucf-101

numpy.core.multiarray failed to import

you should upgrade the numpy version to 1.20.3

Please update code to work with current NVIDIA driver, thanks!

how to get mxnet/python:1.1.0_nccl? I got ``` $ docker build -t extreme-3dmm-docker . Sending build context to Docker daemon 106.4MB Step 1/13 : FROM mxnet/python:1.1.0_nccl manifest for mxnet/python:1.1.0_nccl not found...

get 3d reconstructed faces with facial fine details

When I run `python testBatchModel.py testImages.txt /shared` it warns `sh: 1: ./TestBump: not found`. Is it the reason why my outputs don't have details? How to fix it? ``` $...

get 3d reconstructed faces with facial fine details

Thanks for your reply. Which version of the OpenCV did you use?