Xin Jin

Can you provide some detailed source codes related to this section (i.e., your Fig. 5.)

Some issues when we prepare dataset for generating UV ground truths

2

you wrote "Download human36m dataset from the official website Unpack downloaded zip files into a single folder (which I suggest you name it human36m) and put it under path-to-your-datasets" But,...

when use "--train-sampler RandomIdentitySampler", network can not reach convergence

3

when I use --train-sampler RandomIdentitySampler to perform hardming in the triplets_loss, the model can not reach the convergence (the entropy loss keep constant). Have you met this problem? Should I...

UnicodeDecodeError: 'ascii' codec can't decode byte 0x91 in position 0: ordinal not in range(128)

2

when I load your pre-trained weight,get this problem. Some solutions that provided online didn't work for it, can you provide some suggestions? My setting: Python3.6+PyTorch0.4.1

some questions about "Feature-based Weight Initialization for Classifier."

1

你好，论文里对应的这块内容方便share下实现的code么，想参考下实现方法，感谢，邮箱[email protected]，可私邮。

I want to know the specific parameter settings for HDBSCAN clustering algorithm

1

As your paper demonstrated, HDBSCAN clustering method outperforms DBSCAN, but when I use the commonly-used DBSCAN (from sklearn.cluster import DBSCAN) for clustering, I found that DBSCAN performs better than HDBSCAN...