Gao, Ruiyuan
Gao, Ruiyuan
Hi, In our 272x736 setting, we do not change the map size, which is 200x200. However, the map encoder should be changed to output 34x96 latent, which should be the...
Did you use CFG for inference? The default should be 2, and we use 2.5 to report better results in the table. Another thing I could think of, could you...
Yes, you also need to change them, but the parameters are not correct. First, you should have ```yaml dataset: image_size: [272, 736] augment2d: resize: [[0.5, 0.5]] rotate: null ``` They...
Hi, We have released the 272x736 model with the config. You may want to have a try. Cheers.
Please use "configs/dataset/Nuscenes_map_cache_box.yaml" to configure the dataset, as that from "+exp=224x400". We do not support training with the default one.
I think we have tried the best to do so. Could you double check? Thanks.
We did not strictly verify these options. We did try training image generation by adding the same noise to different views and it does not work. I think the expected...
You may refer to the resources we released here: https://coda-dataset.github.io/w-coda2024/track2/ The only difference is that we use the whole validation set with one sample each in our paper, rather than...
Yes. You may also want to evaluate the control accuracy to check the overall performance.
Just follow the readme, rename the file, and put them in the default path. The code will use it if successfully locates the file. You can also change the file...