Closed ZhouZijie77 closed 1 year ago
Hello, we are creating the database in a style that mimics practical use cases, which requires it to be "outdated". That is why we are creating the database from the first 105 days. Besides, uneven geometric distribution of samples may induce biases when evaluating a place recognition pipeline, so we are keeping each sample in the dataset 1 meter apart from each other.
Thank you!
Hello, thank you for sharing your code. But I don't quite understand why you didn't use all samples to generate the database?