solve merge

configs/local/strategy_generate_config.yaml

@@ -16,14 +16,14 @@ runner:
     compute_with_normal: False
     scan_points_threshold: 10
     overwrite: False
-    seq_num: 15
+    seq_num: 10
     dataset_list:
       - OmniObject3d
   
 datasets:
     OmniObject3d:
-      root_dir: C:\\Document\\Local Project\\nbv_rec\\nbv_reconstruction\\temp
+      root_dir: /data/hofee/nbv_rec_part2_preprocessed
-      from: 0
+      from: 155
-      to: 1 # -1 means end
+      to: 165 # ..-1 means end
 
     

core/global_pts_n_num_pipeline.py

@@ -1,4 +1,5 @@
 import torch
+import time
 from torch import nn
 import PytorchBoot.namespace as namespace
 import PytorchBoot.stereotype as stereotype
@@ -58,7 +59,10 @@ class NBVReconstructionGlobalPointsPipeline(nn.Module):
         return perturbed_x, random_t, target_score, std
 
     def forward_train(self, data):
+        start_time = time.time()
         main_feat = self.get_main_feat(data)
+        end_time = time.time()
+        print("get_main_feat time: ", end_time - start_time)
         """ get std """
         best_to_world_pose_9d_batch = data["best_to_world_pose_9d"]
         perturbed_x, random_t, target_score, std = self.pertube_data(
@@ -117,7 +121,7 @@ class NBVReconstructionGlobalPointsPipeline(nn.Module):
             for seq_idx in range(seq_len): 
                 partial_idx_in_combined_pts = scanned_mask == seq_idx # Ndarray(V), N->V idx mask
                 partial_perpoint_feat = perpoint_scanned_feat[partial_idx_in_combined_pts] # Ndarray(V x Dl)
-                partial_feat = torch.mean(partial_perpoint_feat, dim=0) # Tensor(Dl)
+                partial_feat = torch.max(partial_perpoint_feat, dim=0) # Tensor(Dl)
                 partial_feat_seq.append(partial_feat)
                 scanned_target_pts_num.append(partial_perpoint_feat.shape[0])
                 

runners/strategy_generator.py

@@ -86,13 +86,17 @@ class StrategyGenerator(Runner):
             nrm_path = os.path.join(root,scene_name, "nrm", f"{frame_idx}.npy")
             idx_path = os.path.join(root,scene_name, "scan_points_indices", f"{frame_idx}.npy")
             pts = np.load(pts_path)
-            if pts.shape[0] == 0:
+            if self.compute_with_normal:
-                nrm = np.zeros((0,3))
+                if pts.shape[0] == 0:
-            else:
+                    nrm = np.zeros((0,3))
-                nrm = np.load(nrm_path)
+                else:
-            indices = np.load(idx_path) 
+                    nrm = np.load(nrm_path)
+                nrm_list.append(nrm)
+            
             pts_list.append(pts)
-            nrm_list.append(nrm)
+            
+            indices = np.load(idx_path) 
+            
             scan_points_indices_list.append(indices)
             if pts.shape[0] > 0:
                 non_zero_cnt += 1

utils/reconstruction.py

@@ -75,6 +75,7 @@ class ReconstructionUtil:
         cnt_processed_view = 0
         remaining_views.remove(init_view)
         curr_rec_pts_num = combined_point_cloud.shape[0]
+        drop_output_ratio = 0.4
         
         import time
         while remaining_views:
@@ -84,6 +85,8 @@ class ReconstructionUtil:
             best_covered_num = 0
         
             for view_index in remaining_views:
+                if np.random.rand() < drop_output_ratio:
+                    continue
                 if point_cloud_list[view_index].shape[0] == 0:
                     continue
                 if selected_views: