VLN-DUET
VLN-DUET copied to clipboard
cshizhe
Official implementation of Think Global, Act Local: Dual-scale GraphTransformer for Vision-and-Language Navigation (CVPR'22 Oral).
Hello，When I apply the official code for pretrain ，the val-unseen sr is about 48 of the best ckpt； Then I apply this ckpt to the finetune code ， it easily...
Hi, thank you for opensourcing your work! I have a question about pretraining phase. I'm about to test Duet on R2R, but I wonder when you got the R2R performance...
Hi, thanks for your great work. Can you share how to get the top-down map and the corresponding trajectory? thank you very much
Hello. In the paper, there are some excellent visualizations of the predicted trajectories from top down view in Fig. 8. Is there any code or method to implement these visualizations?
Thank you so much for the great work you do. I reproduced VLN-DUET on the REVERIE datasets according to the file readme.md. But we only get the results for the...
Is pretrained_model_name_or_path missing a path
Hi, could you please provide the object feature extraction code for REVERIE/SOON dataset? It is mentioned in the paper that _"For objects, we use the same ViT on the REVERIE...
Hello, thank you for your excellent work! When I was running bash./scripts/run_reverie.sh to fine-tune the model code, I got a message warning: You are using a model of type bert...
