Wayne Hung

This is likely due to errors from reading images. Please make sure you have the dataset in place

It's simply for generating two sampling sequences for the entire dataset. one for the "real" branch and one for the fake "branch". You can also sample twice from the same...

@ipScore I think it doesn't matter.

If it's unlabeled, then there is no quantitative metric to calculate.

Hi, I'm not experienced with medical datasets so I'm not sure I have good insight for your project. If you can't maintain a good balance between your G and D,...

Hi Sean, I think you can do so by modifying the enum numbers in the label.proto files: https://github.com/waymo-research/waymo-open-dataset/blob/master/waymo_open_dataset/label.proto#L63-L69

Hi, What kind of segmentation mask are you looking for? If you are referring the foreground segmentation mask, feel free to use the provided range image and applying threshold on...

Hi, When calculating L1 metrics, we do not treat any predictions that predict L2 ground truths as false positives. Please refer to https://github.com/waymo-research/waymo-open-dataset/blob/master/waymo_open_dataset/metrics/detection_metrics.cc#L93-L95 for the implementation details. Best, Wayne, on...

Hi, Could you provide your main for loop for reading the images from open datasets?

Hi, We do support returning all precision points in PR curve in the metrics library. Please see https://github.com/waymo-research/waymo-open-dataset/blob/master/waymo_open_dataset/metrics/ops/detection_metrics_ops_test.py#L153 for an example.