Hi,thanks for sharing your great work!
I have ran the instant-ngp on many datasets including fox、lego and my equiretangular dataset in windows system. But I encountered difficulties on an outdoor large-scale dataset. I ran colmap and colmap2nerf and it generated transforms.json successfully. It contains the information of 549 pinhole images.
When I ran the instant-ngp.exe as usual. It got stuck in the dataset loading process. Then I tried to reduce the number of images to 400 by deleting the information in transforms.json. But it still got stuck.
I think this may be due to two reasons.
Images in the dataset have different resolutions.
My GPU performance is too low.
But I can see from the cmd that the GPU's memory and performance have not been fully utilized.
My question is how to find the cause of the program getting stuck and how to solve this problem.
Hi,thanks for sharing your great work! I have ran the instant-ngp on many datasets including fox、lego and my equiretangular dataset in windows system. But I encountered difficulties on an outdoor large-scale dataset. I ran colmap and colmap2nerf and it generated transforms.json successfully. It contains the information of 549 pinhole images. When I ran the instant-ngp.exe as usual. It got stuck in the dataset loading process. Then I tried to reduce the number of images to 400 by deleting the information in transforms.json. But it still got stuck.
I think this may be due to two reasons.
But I can see from the cmd that the GPU's memory and performance have not been fully utilized.
My question is how to find the cause of the program getting stuck and how to solve this problem.