popper0912

> See https://github.com/strawberryfg/DeepModel_hand/blob/master/configuration/DofConstraintId.in. These 21 DoFs are fixed, and so only 26 are tunable by deep network. Here is a brief sketch of correspondence between DoF and joint https://github.com/strawberryfg/DeepModel_hand/blob/master/illustration%20of%20model.jpg. Basically...

> > See https://github.com/strawberryfg/DeepModel_hand/blob/master/configuration/DofConstraintId.in. These 21 DoFs are fixed, and so only 26 are tunable by deep network. Here is a brief sketch of correspondence between DoF and joint https://github.com/strawberryfg/DeepModel_hand/blob/master/illustration%20of%20model.jpg....

> Hi ,Why I flip or rotate your test image,the result is wrong? 你好 问题解决了没 我也遇到了同样的问题 现在不清楚 问题出现在了哪里 期待你的回复！！！

> Is it possible for you to say 'used same pre-processing as Deep prior paper' as in your paper? > Because they did not use GT position as center of...

> [rawHands.tar.gz](https://github.com/xingyizhou/DeepModel/files/1291733/rawHands.tar.gz) > > [rawHands.tar.gz](https://github.com/xingyizhou/DeepModel/files/1291729/rawHands.tar.gz) > hi: > thanks very much for your reply! > I will try the "hands17" dataset(downloading for the moment) > here is my experiments (including...