Open cici19850 opened 3 months ago
Hi, maybe it's just the scene is too hard to colmap to work.
Hi, maybe it's just the scene is too hard to colmap to work.
However, the preprocessing results of the sample data are still very good. I want to try using convert.by to process the raw data of the sample instead of directly using the preprocessed results of the sample data, but the matching rate of the preprocessed results is very low
I did not directly use the sample data mic from nerf_synthetic for training, but instead preprocessed the sample data mic using convert.by. However, the matching rate was very low. After preprocessing 100 sample data using convert.by, only two of them matched. May I ask, what is the reason for this?How can I preprocess data with high matching rates?