search

LMD0311 / PointMamba

[NeurIPS 2024] PointMamba: A Simple State Space Model for Point Cloud Analysis
https://arxiv.org/abs/2402.10739
Apache License 2.0
361 stars 24 forks source link

A question about reported experimental results #4

Closed linchengxing closed 8 months ago

linchengxing commented 8 months ago

Hi, your work is excellent, but I have a question about the reported experimental results. In the item "Training from scratch" in the Table 2 in your paper, according to my understanding, the architecture of "Transformer" and "Point-MAE" should be consistent, but why are there two different experimental results?

LMD0311 commented 8 months ago

Thank you for your interest! Apologies for the confusion. As indicated in the table caption, the signifies the results we replicated. In the "Training from scratch" section, PointMAE represents our reproduced results, while Transformer indicates the results from the Point-BERT paper (which makes a difference in the citation).

LMD0311 commented 8 months ago

I am closing this issue. If you have any more questions, please feel free to reopen it or create new issues. :)