BC03, 1 Sphere, Noise to PCL

Experiment with adding noise to point cloud. See here for visuals on what extra noise means to the point clouds, etc.

Created on June 6|Last edited on June 8

Comment

﻿
OverviewResultsGIFsVisualization of Noise
﻿
Overview(06/06/2022) Running experiments where we are adding noise to the point clouds. Running 5X, after commit.
# 6D flow, but using SVD to extract actions, learning from >= 4DoF demos.
# As of 06/01/2022 this is serving as our 'proposed' method, WITH the
# consistency loss and WITHOUT the separate MLPs for (R,t).
SVD_POINTWISE_6D_EE2FLOW_SVD = dict(
    obs_type='point_cloud',
    act_type='ee2flow',
    encoder_type='pointnet_svd_pointwise_6d_flow',
    method_flow2act='svd',
    use_consistency_loss=True,
    lambda_consistency=0.1,
    scale_pcl_flow=True,
    scale_pcl_val=250,
    separate_MLPs_R_t=False,
    gaussian_noise_PCL=0.001,  # normally 0
)
Should inspect the point cloud visualizations to see how much noise there is added.
Compare with no noise: https://wandb.ai/mooey5775/mixed_media/reports/BC03-4DoF-1-Sphere-PN-SVD-Pointwise-6D-flow-ee2flow-100-demos--VmlldzoyMDM5ODgy﻿
﻿
(06/08/2022) This is strange,  2 runs crashed which is why two do not have the full curves here. HOWEVER, they seem to have actually run to completion on the cluster, so I just need to use my own script which should load in the complete files?
Results﻿
6D Flow, SVD, Pointwise (Noise 0.001)5
﻿
GIFs﻿
﻿
Visualization of NoiseLeft: no noise. Right: the noise scale of 0.001.
﻿
﻿

Add a comment