search

ChikaYan / d2nerf

Apache License 2.0
185 stars 14 forks source link

Adaptivity to other dynamic datasets #20

Closed ntaquan0125 closed 5 months ago

ntaquan0125 commented 7 months ago

Dear authors,

Thank you for your excellent work.

I have tested the method on several outdoor scenes involving high ego-motion; however, the results are not entirely satisfactory. Do you believe there might be a gap in adapting to this type of data, or are there adjustments I could make to achieve better separation in such scenarios?

I also played with the hyperparameters but this is the best I can get.

d2nerf_kitti

ChikaYan commented 7 months ago

Hi,

Thx for your interest in our work. Decoupling of outdoor scenes might be quite challenging due to global brightness changes. You might be able to get better results by tweaking the skewness hyperparameters, but know that this is a relatively old work and you might want to try some newer methods such as https://omnimatte-rf.github.io/ first to see if it gives better results.

Best