Some works may not be awesome？

[joe-zxh]

ARGAN: Fast Converging GAN for Animation Style Transfer

This works seems simply copy the work from AnimeGAN. The network architecture and losses design are the same with the AnimeGAN.

Cartoonize Images using TinyML Strategies with Transfer Learning
This work seems simply convert the model to a quantilized 8-bit model using TFLite. It may be a trival work.

[SerialLain3170]

Hello,

I am sorry for the late reply, and thank you for the suggestion. Those are the good points.

• As for the first paper, what you mentioned is true. No obvious differences between the paper and AnimeGAN can be found.
• When it comes to the second paper, although making the network light is very important for the industrial application, it may be trivial as an academic work because these are less novel points in the method.

So, I have a plan to remove these two papers from the list. But, I will not remove soon because there would be novel points that we do not notice.