We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
作者你好,我再使用你的参数来对MiddleBurySet进行推理时,发现对很多样本的插帧存在细节模糊等问题,这个你有研究过如何产生的吗。我跑了多个模型,其实都存在这个问题,很少能够非常好的保留出细节
The text was updated successfully, but these errors were encountered:
Hi,我感觉这个数据集本身好像也不是特别清晰 然后用 vgg loss 训练的模型,清晰度会高一些,l1 loss 倾向于模糊
Sorry, something went wrong.
No branches or pull requests
作者你好,我再使用你的参数来对MiddleBurySet进行推理时,发现对很多样本的插帧存在细节模糊等问题,这个你有研究过如何产生的吗。我跑了多个模型,其实都存在这个问题,很少能够非常好的保留出细节
The text was updated successfully, but these errors were encountered: