Lue Fan

Validation split. Single frame; lidar-only; No ensemble; Just for reference. ``` OBJECT_TYPE_TYPE_VEHICLE_LEVEL_1: [mAP 0.751252] [mAPH 0.746433] OBJECT_TYPE_TYPE_VEHICLE_LEVEL_2: [mAP 0.666054] [mAPH 0.661673] OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_1: [mAP 0.800718] [mAPH 0.72117] OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_2: [mAP 0.723756] [mAPH...

Detailed Performance of FSD

4

Performance of `fsd_waymoD1_1x_submission.py` (test set): [submission](https://waymo.com/open/challenges/entry/?timestamp=1665211204047769&challenge=DETECTION_3D&emailId=1cb154ab-1558) Performance of `fsd_waymoD1_1x.py` (validation set): ``` OBJECT_TYPE_TYPE_VEHICLE_LEVEL_1: [mAP 0.792293] [mAPH 0.787888] OBJECT_TYPE_TYPE_VEHICLE_LEVEL_2: [mAP 0.704975] [mAPH 0.700932] OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_1: [mAP 0.825951] [mAPH 0.773279] OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_2: [mAP 0.739378]...

Question about NCECriterion. py

3

https://github.com/HobbitLong/CMC/blob/58d06e9a82f7fea2e4af0a251726e9c6bf67c7c9/NCE/NCECriterion.py#L30 I think the purpose of using NCE is to avoid expensive summation over entire vector in softmax. But in your implementation, there is still summation over entire `log_D0` which...

How to understand find_coeffs function in save_homograhpy.py?

1

`res = numpy.dot(numpy.linalg.inv(A.T * A) * A.T, B)` What do A and B denote? And how to understand the operation above? Your paper is excellent, but I find the snippet...

Experimental results of given configs.

4

Nice work! May I ask if it is convenient for you to provide the basic results of the three given configs (CenterPoint 50/100/200 ) for reference? And since the point...