Open-PIFuhd icon indicating copy to clipboard operation
Open-PIFuhd copied to clipboard

This is an implementation of PIFuhd based on Pytorch

Results 20 Open-PIFuhd issues
Sort by recently updated
recently updated
newest added

how to get ground truth of Normal ??

what is the folder structure required for inference , I have provided "./Garment/render_gen_1024_test/test.png" , it is showing test size as 0 for --input_dir = './Garment/render_gen_1024_train/' , have anyone used it...

> follow the dataset I write. https://github.com/lingtengqiu/Open-PIFuhd/blob/master/engineer/datasets/RPDataset.py, you could simply set camera matrix when u test Thank you for your reply. I want to ask you one more question. During...

os.fork was called ... , i don't know what is the issue I am trying to inference/test this project , i have passed single img at '/Garment/render_gen_1024_test/test.png' and loaded the...

作者您好,感谢您对于pifuhd工作的实现。 我有一个关于渲染数据集的问题。 请问RENDER_NORMAL是必要的训练数据吗?

如果我想要测试一张真实的图片,是否需要编写额外的测试脚本

The entire github code has been run, and the resulting obj model is as follows: The obj model generated by bash. / scripts/test_pfhd_coarse.sh: ![image](https://github.com/lingtengqiu/Open-PIFuhd/assets/137875693/3042458f-e6b5-47c5-9c29-c10b55560dfd) ![image](https://github.com/lingtengqiu/Open-PIFuhd/assets/137875693/375ccce3-4455-4bee-8830-00a01011bd00) The obj model generated by...

Traceback (most recent call last): File "./tools/train_pifu.py", line 56, in train_dataloader = build_dataloader(train_data_set,cfg,args) File "/home/gaojunxiao/train_data/Open-PIFuhd/./utils/dataloader.py", line 33, in build_dataloader return get_loader(dataset,batch_size = batch_size,num_workers = num_workers,collect_fn = collect_fn,shuffle=shuffle,worker_init_function=worker_init_function) File "/home/gaojunxiao/train_data/Open-PIFuhd/./utils/dataloader.py", line...

![image](https://user-images.githubusercontent.com/48088535/203599063-82498cf7-5f33-432a-b9cc-85a9f5246285.png)

您好,我下载了一些renderpeople数据,但是不知道Garment文件具体结构以及数据预处理碰到一些问题,不知您是否方便交流一下,我的邮箱是[email protected],万分感谢!