search

runnanchen / CLIP2Scene

Other
153 stars 6 forks source link

Questions about the results #7

Open Yu-Jiawei opened 1 year ago

Yu-Jiawei commented 1 year ago

Why is the result of Table 1 comparison in the paper inconsistent with the original SLidR paper?

runnanchen commented 1 year ago

Why is the result of Table 1 comparison in the paper inconsistent with the original SLidR paper?

SLidR uses MinkUnet as the point backbone, while we use the SPVCNN as the point backbone because our machine did not support the minkowski engine at that time. We will supplement the MinkUnet benchmark soon. Thanks.

Yu-Jiawei commented 1 year ago

Why is the result of Table 1 comparison in the paper inconsistent with the original SLidR paper?

SLidR uses MinkUnet as the point backbone, while we use the SPVCNN as the point backbone because our machine did not support the minkowski engine at that time. We will supplement the MinkUnet benchmark soon. Thanks.

Thanks for your relpy, but I still have a question. Compared with the results in the SLidR paper, why does using SPVCNN as the backbone have a greater improvement in the nuscenes dataset, but a decline in the kitti dataset?