chenyu-yang-2000 / elegant Goto Github PK

I came across your Upwork job post.
If you would prefer, I am always available for a call to discuss further and get started with the work right away.
[email protected]
https://join.skype.com/invite/scT8rfuam2qI
Pls share your contact info.

老师您好，用自己的图片运行demo,输出的结果是被裁减后的图片，请问如何修改可以输出没有被裁减的图片呢？

Thank you for your outstanding work. I used MT datasets to produce the following results just through PGT. May I ask how this happened? What are the results you produce with PGT ?

How can I finetune the model? I am trying to load the model and start training but the seg folder is missing in my dataset. How can I create the seg folder and the images inside?

Can we chat on Telegram?
What is your Telegram ID?

I am really appreciate your code, and I am interested in this code. I am studying EleGANt and want to give some changes. I found that this EleGANt model cannot recognize 2 person. So I want to get your advice. How can I change the code to recognize two person? WHICH CODE I NEED TO STUDY AND REFORM? THANK YOU.

Thank you for your outstanding work. How can I obtain images generated by PGT

你好，请问我要如何转换为onnx模型？

Can the author provide a training dataset? especially in numpy format

您好:

不好意**請問作者face detection設置up ratio, down ratio跟width ratio的用意? 以及這個參數數值是如何訂的呢?

因為我將其他人臉資料集作為測試集時，發現經過這些ratio反而會對face detection產生反效果。例如本來抓到的bounding box是正常的，但經過ratio校正後bounding box偵測到的人臉區塊反而會偏移許多，如下圖所示人臉的部分反而抓的不多。

因此想詢問作者這些參數該如何設置，或是可以去除這些參數設置嗎? 目前我是拿video的人臉資料集套github方法。首先是遇到這個問題，但不知道該怎麼設置參數比較好(或是不知道是否可以拔除這個部份)，也不知道參數設置的用意。

謝謝作者的回覆

您好，感谢您的elegant工作，
想询问下我在调整分辨率到512后训练batch_size=1（A100）需要52gb的显存，2.5小时每epoch
在训练batch_zize=2时显存爆掉
然后希望使用多卡训练，该如何使用呢？

大佬，你好。想问一下，demo视频里面交互的要怎么运行起来。
不知道是不是调用 inference里面的 interface_transfer

Hi, thanks for great work, I have a question as follows:
how to get global shade control result as Fig 1 shows in the paper? which shade can be adjust from light to heavy?