Lcon7
Lcon7
@memorylorry SegmentationClassAug can be downloaded from here: https://www.dropbox.com/s/oeu149j8qtbs1x0/SegmentationClassAug.zip?dl=0 I found this on the Internet
Hi, sorry for the late reply... We will push the codes for nuscenes soon
That is a deprecated function and will no longer be used. We will delete this part in the future. Sorry for the confusion.
Hi, Thanks for the question. We use the 500/125 human annotated samples, plus the remaining 3712-500/125 samples (which are annotated by the MTrans auto labeler) to train the PointPillars/PointRCNN from...
Those rows are the results for PointPillars/PointRCNN trained with 500/125 human annotations only, no pseudo labels. We use the "500f"/"125f" to denote how many human annotations are required for these...
Yes, that is correct. Thanks.
We removed labels with too much truncation at [here](https://github.com/Cliu2/MTrans/blob/52e1b81def3ef19b4dded9d1f0faa5008b33e4ea/datasets/kitti_detection.py#L95). The target objects have way too few LiDAR points to generate a good enough pseudo label, and therefore are omitted, resulting...
They are simply not supervised, no loss is calculated for those objects. Empty txt can be used.
In our experiments, we just replace the label_2 files with the generated pseudo labels. Albeit the results could vary from time to time due to the randomness of different environments,...