Closed xiaofeiso closed 1 year ago
Hi, I have the exact same issue. Ill post here if i find what's going on. Update me if you find it too please!
I do not have this issue with the tutorials, only when using either the tutorial example with my own .pcd files or when using gpd_ros while subscribed to a pointcloud2 message.
I noticed the format of my PCD clouds is a bit different than the ones in the tutorials:
my pcd:
# .PCD v0.7 - Point Cloud Data file format VERSION 0.7 FIELDS x y z rgb SIZE 4 4 4 4 TYPE F F F U COUNT 1 1 1 1 WIDTH 3110 HEIGHT 1 VIEWPOINT 0 0 0 1 0 0 0 POINTS 3110 DATA ascii
Tutorial's PCD:
# .PCD v.7 - Point Cloud Data file format VERSION .7 FIELDS x y z rgb SIZE 4 4 4 4 TYPE F F F I COUNT 1 1 1 1 WIDTH 4467 HEIGHT 1 POINTS 4467 DATA ascii
OK, if I find a solution, I will reply it.
---Original--- From: @.> Date: Fri, Nov 25, 2022 06:28 AM To: @.>; Cc: @.**@.>; Subject: Re: [atenpas/gpd] [pcl::NormalEstimationOMP::compute] input_ isempty! (Issue #131)
Hi, I have the exact same issue. Ill post here if i find what's going on. Update me if you find it too please!
I do not have this issue with the tutorials, only when using either the tutorial example with my own .pcd files or when using gpd_ros while subscribed to a pointcloud2 message.
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
Ho wow, look at that!
workspace: -1.00 1.00 -1.00 1.00 -1.00 1.00 candidate_workspace: -1.00 1.00 -1.00 1.00 -1.00 1.00
My cloud happened to fall outside of this region. I changed the values in order to have my cloud inside of the region. Now it's not generating the same output. Normals are able to be compputed but i dont have a result yet (maybe i bugged something else while fideling into stuff...
i confirm it fixed the issue for me.
Be aware though, if you run the example, the window is made to visualize an object located at 0 0 0. It's normal that you have to change your point of view in the 3D viewer in order to see what's going on.
ok, the error is solved,but nomal ,candidate grasp not generated, Do you mean to modify camera_position?
well in the config files i modified the fields of workspace and candidate_workspace to dimensions that fit my use case. it's hard to get a definitive answer since i do not know the parameters of your project.
When I try my own point cloud,I run ./detect_grasps ../cfg/eigen_params.cfg,/home/xiaofeisong/Desktop/f/code/our_dataset/luosuan/google_512k/nontextured.pcd
Failed to find match for field 'rgba'. Loaded point cloud with 76290 points ============ HAND GEOMETRY ====================== finger_width: 0.01 hand_outer_diameter: 0.12 hand_depth: 0.06 hand_height: 0.02 init_bite: 0.01
============ PLOTTING ======================== plot_normals: true plot_samples false plot_candidates: true plot_filtered_candidates: false plot_valid_grasps: false plot_clustered_grasps: false plot_selected_grasps: true
============ CLOUD PREPROCESSING ============= voxelize: true voxel_size: 0.003 remove_outliers: false workspace: -1.00 1.00 -1.00 1.00 -1.00 1.00 sample_above_plane: false normals_radius: 0.030 refine_normals_k: 0
============ CANDIDATE GENERATION ============ num_samples: 30 num_threads: 4 nn_radius: 0.01 hand axes: 2 num_orientations: 8 num_finger_placements: 10 deepen_hand: true friction_coeff: 20.00 min_viable: 6
============ GRASP IMAGE GEOMETRY =============== volume width: 0.1 volume depth: 0.06 volume height: 0.02 image_size: 60 image_num_channels: 15
NET SETUP runtime: 0.0433275 ============ CLASSIFIER ====================== model_file: weights_file: ../models/lenet/15channels/params/ batch_size: 1
============ CANDIDATE FILTERING ============= candidate_workspace: -1.00 1.00 -1.00 1.00 -1.00 1.00 min_aperture: 0.0000 max_aperture: 0.0850
============ CLUSTERING ====================== min_inliers: 0
Processing cloud with 76290 points. Voxelized cloud: 0 Calculating surface normals ... numthreads: 4 [pcl::NormalEstimationOMP::compute] input is empty! runtime(computeNormals): 0.0000 camera: 0, #indices: 0, #normals: 0 Calculated 0 surface normals in 0.0000s (mode: OpenMP). Reversing direction of normals that do not point to at least one camera ... reversed 0 normals runtime (reverse normals): 3.123e-06 Plotting normals for different camera sources