Skip to content

Replication of BSRGAN results #42

Description

@Magauiya

Dear authors,

I am more than interested in your work and would very much appreciate if you answer the following questions:

  • Have you used the blindsr.degradation_bsrgan script to generate test dataset? I inferenced BSRNet and RRDB from type-I to type-3 (of proposed DIV2K4D) and achieved almost the same results as reported in the paper. However, results on type-4 (generated by abovementioned script) led to significantly worse results: BSRNet 24.40 +- 0.14 dB (over 10 realizations) and RRDB: 22.50 +- 0.14 dB. Metrics are calculated in Y channel.
  • Do you use the same script to generate a training dataset? Also, are the training parameters (learning rate, patch shape (72x72), num. of iterations, etc.) of BSRNet the same as BSRGAN?

Sincerely,
Magauiya

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions