About regularization loss

[shoukna]

Hi, In the paper, regularization is used to drive the separation of the texture and the shape representations. I'm doing a similar work, but I don't have frameC to do regularization. If there is no regularization, will the separation result be poor? And are there other alternatives to regularization? Thank you~

[12dmodel]

We expect regularization to help. If it's not available, it might be worth trying to train with more frames. See figure 1-13d of my thesis( https://dspace.mit.edu/handle/1721.1/122560) where we trained the network on not just 2-frame, but with 5-frame sequence as input.

On Sat, May 8, 2021 at 1:46 AM shoukna @.***> wrote:

Hi, In the paper, regularization is used to drive the separation of the texture and the shape representations. I'm doing a similar work, but I don't have frameC to do regularization. If there is no regularization, will the separation result be poor? And are there other alternatives to regularization? Thank you~

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/12dmodel/deep_motion_mag/issues/21, or unsubscribe https://github.com/notifications/unsubscribe-auth/AANUQV7R5KFUM7ACFQPXUGLTMT27DANCNFSM44MPTMTQ .

[shoukna]

Thank a lot! But the main problem is that it is more difficult to synthesize data, I can't directly use the public dataset you provided, can you share the code of your synthetic data?

[pratyushganguly]

Hii have you been able to execute the codes and got an output ? If yes, how ? kindly help I'm new to programming

just elaborating the steps would be enough, I am not able to clearly understand the given tutorial. thanks,