leilai125

Supervision loss in code and paper

3

Hi, thanks for your work. It's very interesting. I realized that during supervision part, the loss function is different between the code and paper, right? In the paper, the loss...

About information matrix

5

Hi, thanks for your sharing. I noticed that you set Information matrix for odometry edges, = 500.0 0.0 0.0 500.0 0.0 500.0 and Information matrix for loop closure edges, =...

About code of GenAD

4

Hi, would you please release the inference code and trained model of GenAD? Thank you very much!

Question about the training time?

1

Hi, Thanks for the awesome work! From the paper, all models are trained with 8 80GB A100 GPUs, which takes around 16 hours. But why did the appendix on huggingface...

AttributeError: type object 'StableVideoDiffusionPipeline' has no attribute 'save_lora_weights'

5

Thanks for your work! I am trying to run train_svd_lora.py but returned this error. It seems we can't fine-tune the SVD using Lora. Should I add the 'save_lora_weights' function to...