search

AllenXiangX / SnowflakeNet

(TPAMI 2023) Snowflake Point Deconvolution for Point Cloud Completion and Generation with Skip-Transformer
MIT License
142 stars 16 forks source link

L1 chamfer distance for PCN dataset is 5.07*10^(-3), which is better than 7.21*10^(-3) in the paper, is this correct? #9

Closed zichang06 closed 2 years ago

zichang06 commented 2 years ago

Hi, I trained the model based on the PCN dataset. The average L1 chamfer distance is 5.1510^(-3). The pre-trained model you provided is 5.0710^(-3), which is better than the data 7.21*10^(-3) in the paper. Except for the dataset path, I did not modify your code. Is this result correct?

AllenXiangX commented 2 years ago

Hi, I think it's probably because the data you used for testing is not right. 5.0710^(-3) is too low.

zichang06 commented 2 years ago

Hi, I think it's probably because the data you used for testing is not right. 5.0710^(-3) is too low.

Thank you~ The dataset accesseed in my code was indeed wrong. I have already fixed this problem and get the same result in your paper. (๑•̀ㅂ•́)و✧