bevfusion icon indicating copy to clipboard operation
bevfusion copied to clipboard
verified publisher icon mit-han-lab

can i train bevfusion with kitti dataset

Open Baron6977 opened this issue 2 years ago • 11 comments

Thank you for the author's contribution ：） Can I train bevfusion with kitti dataset？？？？

Baron6977 avatar Mar 21 '23 07:03 Baron6977

Yes it is also possible, but probably you will have to implement the dataloaders for KITTI by yourself. There might be some references from mmdet3d that could be very helpful, but I am not very familiar with it.

kentang-mit avatar Mar 23 '23 20:03 kentang-mit

@Baron6977 Hello, I meet the same problem with you, may I ask if you were successful?

xxp912 avatar Apr 07 '23 08:04 xxp912

Hello, I meet the same problem with you, may I ask if you were successful?

chenzb0712 avatar Aug 10 '23 01:08 chenzb0712

Hello, I meet the same problem with you, may I ask if you were successful?

Lcl159 avatar Sep 26 '23 13:09 Lcl159

@Baron6977 Hello, I meet the same problem with you, may I ask if you were successful?

i can

liuyuansheng624 avatar Nov 30 '23 04:11 liuyuansheng624

Hello, I meet the same problem with you, may I ask if you were successful?

sure

liuyuansheng624 avatar Nov 30 '23 04:11 liuyuansheng624

@liuyuansheng624 你好，请问您是怎么训练kitti数据集的，方便分享一下吗？我怎么联系您？你可以[email protected] 联系我.

guoningwang avatar Dec 03 '23 07:12 guoningwang

@Baron6977 Hello, I meet the same problem with you, may I ask if you were successful?

i can

Hello! How about the performance in the KITTI (compared to the Voxel-RCNN)? Would you share the evaluation? Thanks!

weilli avatar Dec 17 '23 14:12 weilli

you can  wechat 13731082126 for me

发自我的iPhone

------------------ Original ------------------ From: weilli @.> Date: Sun,Dec 17,2023 10:08 PM To: mit-han-lab/bevfusion @.> Cc: liuys @.>, Mention @.> Subject: Re: [mit-han-lab/bevfusion] can i train bevfusion with kitti dataset(Issue #349)

liuyuansheng624 avatar Dec 17 '23 14:12 liuyuansheng624

Hello, I meet the same problem with you, may I ask if you were successful?

hello! Have you succeeded, please！

c-yyyy avatar Apr 09 '24 11:04 c-yyyy

加我微信13731082126

发自我的iPhone

------------------ Original ------------------ From: chen @.> Date: Tue,Apr 9,2024 7:06 PM To: mit-han-lab/bevfusion @.> Cc: liuys @.>, Mention @.> Subject: Re: [mit-han-lab/bevfusion] can i train bevfusion with kitti dataset(Issue #349)

Hello, I meet the same problem with you, may I ask if you were successful?

hello! Have you succeeded, please！

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.***>

liuyuansheng624 avatar Apr 09 '24 11:04 liuyuansheng624