Question about table3.

[johnnylu305]

Hi, thank you for the code. I wonder why the psnr (29.26) in this github is slightly lower than the table3 (30.78) in the paper. Do I misunderstand something?

[WangFeng18]

Table 3 in our paper is for the mean PSNR over all scenes, we have stated in the caption of Table 3. The 29.26 number you stated refers to the performance of flame_salmon scene with the MixVoxels-T schedule.

[johnnylu305]

Thank you for the respond. I want to make sure my understanding is correct. The PSNR of Dy-NeRF and NeRF-T in Table 3 are reported from their paper and they are PSNR of flame_salmon scene according to StreamRF. (AlgoHunt/StreamRF#1)

[WangFeng18]

According to the original paper, we infer that the PSNRs in the DyNeRF is evlauated on all 300 frame videos, I can not find any hints in the original paper that the PSNR number is reported in the flame salmon scene

[johnnylu305]

Really thank you for the help. I also check these papers again. I think you are right.

[johnnylu305]

I recently revisit this post. Perhaps, it is helpful for others.