How to color the src and ref point clouds with learned features?

[MR-hyj]

The input train_data in model contains train_data['points_src'] of 8x717x6 and train_data['points_ref"] of 8x717x6. I extracted the following feature for each sample as the author defined during learning:

# feature returned by sample_and_group_multi()
feature_before_dim = {
    'xyz'  :     # an array of (717, 1, 3),
    'dxyz':     # an array of (717, 64, 3),
    'ppf':       # an array of (717, 64, 4)
}

# feature returned by FeatExtractionEarlyFusion.forward()
feature_after_dim     # an array of (717, 96)

Here's the problem. How either of the above two features help me color the 717 points in src clouds? It is expected that the points with similar features should share the same color. In other words, given n colors (blue, red, etc..), how to design a hash function f such that f(feat) -> {0, 1, 2, ..., n-1} maps the input feat to a particular color

[MR-hyj]

I designed my own local feature of point clouds. Now I'm working on the comparison between other features and mine via coloring points with mapped color from learned features.

I've tried to set the feat_dim=1 in arguments.py, but it did'nt work as anticipated.

[yewzijian]

Hi, if I understand correctly, you are trying to visualize the high dimensional features on the point cloud using color.

• If you are discretizing it into n colors, you need to cluster the features into n clusters, and assign a different color to each of the clusters.
• If you are using a continuous RGB range, you'll first need to map the high dimensional feature spaces into three dimensions, e.g. using PCA or tSNE.

Zi Jian

[MR-hyj]

el

I will try PCA. Thank you. You've been very helpful