merge

README.md

@@ -1,12 +1,12 @@
 ---
-title: Generalizable-HOI-Denoising
-emoji: 🐠
+title: Test
+emoji: 💻
 colorFrom: indigo
 colorTo: green
 sdk: gradio
-sdk_version: 4.36.0
+sdk_version: 4.17.0
 app_file: app.py
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -12,8 +12,6 @@ import shutil
 # from gradio_inter.predict_from_file import predict_from_file
 from gradio_inter.create_bash_file import create_bash_file
 
-from sample.reconstruct_data_taco import reconstruct_from_file
-
 def create_temp_file(path: str) -> str:
     temp_dir = tempfile.gettempdir()
     temp_folder = os.path.join(temp_dir, "denoising")
@@ -43,98 +41,16 @@ def predict(file_path: str):
     
     res_file_path = "/tmp/denoising/save/predicted_infos_seed_0_tag_20231104_017_jts_spatial_t_100__st_0.npy"
     
-    saved_path = reconstruct_from_file(temp_file_path)
-    
-    return saved_path
-
-def create_demo():
-
-    USAGE = """# GeneOH Diffusion: Towards Generalizable Hand-Object Interaction Denoising via Denoising Diffusion
-    **[Project](https://meowuu7.github.io/GeneOH-Diffusion/) | [Paper](https://openreview.net/pdf?id=FvK2noilxT) | [Github](https://github.com/Meowuu7/GeneOH-Diffusion)**
-    ## Input data format
-    Currently, the demo accepts a `.pkl` file containing an hand-object sequence organized as the following format:
-    ```python
-        {
-            "hand_pose": numpy.ndarray(seq_length, 48), # MANO pose at each frame
-            "hand_trans": numpy.ndarray(seq_length, 3), # hand global translation at each frmae
-            "hand_shape": numpy.ndarray(10), # MANO shape coefficients
-            "hand_verts": numpy.ndarray(seq_length, 778, 3), # MANO hand vertices
-            "hand_faces": numpy.ndarray(1538, 3), # MANO hand faces
-            "obj_verts": numpy.ndarray(seq_length, num_obj_verts, 3), # object vertices at each frame
-            "obj_faces": numpy.ndarray(num_obj_faces, 3), # object faces
-            "obj_pose": numpy.ndarray(seq_length, 4, 4), # object pose at each frame
-        }
-    ```
-    We provide an example [here](https://drive.google.com/file/d/17oqKMhQNpRqSdApyuuCmTrPkrFl0Cqp6/view?usp=sharing). **The demo is under developing and will support more data formats in the future.**
-    
-    
-    ## To run the demo, 
-    1. Upload a `pickle` file to the left box by draging your file or clicking the box to open the file explorer.
-    2. Clik the `Submit` button to run the demo.
-    3. The denoised sequence will be output as a `.npy` file and can be downloaded from the right box. 
-    
-    Since the model runs on CPU currently, the speed is not very fast. For instance, it takes abount 1200s to process the [example](https://drive.google.com/file/d/17oqKMhQNpRqSdApyuuCmTrPkrFl0Cqp6/view?usp=sharing) mentioned above which contains 288 frames. Please be patient and wait for the result. 
+    return res_file_path
     
-    To run the model faster, please visit our [github repo](https://github.com/Meowuu7/GeneOH-Diffusion), follow the instructions and run the model on your own server or local machine.
-    
-    ## Output data format 
-    The output is a `.npy` file containing the denoised sequence organized as the following format:
-    ```python
-        {
-            "predicted_info": {
-                "targets": numpy.ndarray(seq_length, num_mano_joints, 3), # input MANO joints
-                "outputs": numpy.ndarray(seq_length, num_mano_joints, 3), # denoised MANO joints
-                "obj_verts": numpy.ndarray(seq_length, num_obj_verts, 3), # object vertices at each frame
-                "obj_faces": numpy.ndarray(num_obj_faces, 3), # object faces
-                ... # others
-            }
-            "bf_ct_verts": numpy.ndarray(seq_length, 778, 3), # denoised MANO vertices
-            "bf_ct_rot_var": numpy.ndarray(seq_length, 3), # denoised MANO global rotation coefficients
-            "bf_ct_theta_var": numpy.ndarray(seq_length, 45), # denoised MANO global pose coefficients
-            "bf_ct_beta_var": numpy.ndarray(1, 10), # denoised MANO shape coefficients
-            "bf_ct_transl_var": numpy.ndarray(seq_length, 3), # denoised hand global translation
-        }
-    ```
-    The corresponding output file of the [example](https://drive.google.com/file/d/17oqKMhQNpRqSdApyuuCmTrPkrFl0Cqp6/view?usp=sharing) mentioned above can be downloaded [here](https://drive.google.com/file/d/1Ah-qwV6LXlOyaBBe0qQRu1lN-BpKt2Y3/view?usp=sharing).
-    """
-
-
-    with gr.Blocks() as demo:
-
-        gr.Markdown(USAGE)
 
-        # # demo = 
-        # gr.Interface(
-        #     predict,
-        #     # gr.Dataframe(type="numpy", datatype="number", row_count=5, col_count=3),
-        #     gr.File(type="filepath"),
-        #     gr.File(type="filepath"),
-        #     cache_examples=False
-        # )
+demo = gr.Interface(
+    predict,
+    # gr.Dataframe(type="numpy", datatype="number", row_count=5, col_count=3),
+    gr.File(type="filepath"),
+    gr.File(type="filepath"),
+    cache_examples=False
+)
 
-
-        input_file = gr.File(type="filepath")
-        output_file = gr.File(type="filepath")
-        
-        gr.Interface(
-            predict,
-            # gr.Dataframe(type="numpy", datatype="number", row_count=5, col_count=3),
-            input_file,
-            output_file,
-            cache_examples=False
-        )
-        
-        inputs = input_file
-        outputs = output_file
-        gr.Examples(
-            examples=[os.path.join(os.path.dirname(__file__), "./gradio_inter/20231104_017.pkl"), os.path.join(os.path.dirname(__file__), "./gradio_inter/20231104_010.pkl")],
-            inputs=inputs,
-            fn=predict,
-            outputs=outputs,
-        )
-
-    return demo
-    
 if __name__ == "__main__":
-    demo = create_demo()
     demo.launch()

data_loaders/humanml/data/dataset_ours_single_seq.py

@@ -7252,7 +7252,7 @@ class GRAB_Dataset_V19_HHO(torch.utils.data.Dataset): # GRAB datasset #
             flat_hand_mean=True,
             side='left',
             # mano_root=self.mano_path, # mano_root #
-            mano_root=self.mano_path,
+            mano_root="/home/hlyang/HOI/HOI/manopth/mano/models",
             ncomps=45,
             use_pca=False,
             # center_idx=0

gradio_inter/20231104_010.pkl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:2224d2863679dee3f24820538be025cf4c11bb9117b99797f83ca267741a9642
-size 2978083

gradio_inter/20231104_017.pkl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:6c432a154c16c0b135b73162e4d822516fbe3c1d36933f1a54b791cbfa9365ed
-size 3215727

gradio_inter/predict_from_file.py

@@ -199,7 +199,7 @@ def main():
     st_idxes = list(range(0, num_ending_clearning_frames, nn_st_skip))
     if st_idxes[-1] + num_cleaning_frames < nn_frames:
         st_idxes.append(nn_frames - num_cleaning_frames)
-    # st_idxes = [st_idxes[0]]
+    st_idxes = [st_idxes[0]]
     print(f"st_idxes: {st_idxes}")
     
     

sample/reconstruct_data_taco.py

@@ -51,15 +51,15 @@ def get_penetration_masks(obj_verts, obj_faces, hand_verts):
 def get_optimized_hand_fr_joints_v4_anchors(joints, base_pts, tot_base_pts_trans, tot_base_normals_trans, with_contact_opt=False, nn_hand_params=24, rt_vars=False, with_proj=False, obj_verts_trans=None, obj_faces=None, with_params_smoothing=False, dist_thres=0.005, with_ctx_mask=False):
   # obj_verts_trans, obj_faces
   joints = torch.from_numpy(joints).float() #  # joints 
-  # base_pts = torch.from_numpy(base_pts).float() #  # base_pts 
+  base_pts = torch.from_numpy(base_pts).float() #  # base_pts 
   
   if nn_hand_params < 45:
     use_pca = True
   else:
     use_pca = False
   
-  # tot_base_pts_trans = torch.from_numpy(tot_base_pts_trans).float()
-  # tot_base_normals_trans = torch.from_numpy(tot_base_normals_trans).float()
+  tot_base_pts_trans = torch.from_numpy(tot_base_pts_trans).float()
+  tot_base_normals_trans = torch.from_numpy(tot_base_normals_trans).float()
   ### start optimization ###
   # setup MANO layer
   # mano_path = "/data1/xueyi/mano_models/mano/models"
@@ -139,48 +139,48 @@ def get_optimized_hand_fr_joints_v4_anchors(joints, base_pts, tot_base_pts_trans
   # )
   
   # 
-  # dist_joints_to_base_pts = torch.sum(
-  #   (joints.unsqueeze(-2) - tot_base_pts_trans.unsqueeze(1)) ** 2, dim=-1 # nf x nnjoints x nnbasepts #
-  # )
+  dist_joints_to_base_pts = torch.sum(
+    (joints.unsqueeze(-2) - tot_base_pts_trans.unsqueeze(1)) ** 2, dim=-1 # nf x nnjoints x nnbasepts #
+  )
   
-  # nn_base_pts = dist_joints_to_base_pts.size(-1)
-  # nn_joints = dist_joints_to_base_pts.size(1)
+  nn_base_pts = dist_joints_to_base_pts.size(-1)
+  nn_joints = dist_joints_to_base_pts.size(1)
   
-  # dist_joints_to_base_pts = torch.sqrt(dist_joints_to_base_pts) # nf x nnjoints x nnbasepts #
-  # minn_dist, minn_dist_idx = torch.min(dist_joints_to_base_pts, dim=-1) # nf x nnjoints #
+  dist_joints_to_base_pts = torch.sqrt(dist_joints_to_base_pts) # nf x nnjoints x nnbasepts #
+  minn_dist, minn_dist_idx = torch.min(dist_joints_to_base_pts, dim=-1) # nf x nnjoints #
   
-  # nk_contact_pts = 2
-  # minn_dist[:, :-5] = 1e9
-  # minn_topk_dist, minn_topk_idx = torch.topk(minn_dist, k=nk_contact_pts, largest=False) # 
-  # # joints_idx_rng_exp = torch.arange(nn_joints).unsqueeze(0) == 
-  # minn_topk_mask = torch.zeros_like(minn_dist)
-  # # minn_topk_mask[minn_topk_idx] = 1. # nf x nnjoints #
-  # minn_topk_mask[:, -5: -3] = 1.
-  # basepts_idx_range = torch.arange(nn_base_pts).unsqueeze(0).unsqueeze(0)
-  # minn_dist_mask = basepts_idx_range == minn_dist_idx.unsqueeze(-1) # nf x nnjoints x nnbasepts
-  # # for seq 101
-  # # minn_dist_mask[31:, -5, :] = minn_dist_mask[30: 31, -5, :]
-  # minn_dist_mask = minn_dist_mask.float()
+  nk_contact_pts = 2
+  minn_dist[:, :-5] = 1e9
+  minn_topk_dist, minn_topk_idx = torch.topk(minn_dist, k=nk_contact_pts, largest=False) # 
+  # joints_idx_rng_exp = torch.arange(nn_joints).unsqueeze(0) == 
+  minn_topk_mask = torch.zeros_like(minn_dist)
+  # minn_topk_mask[minn_topk_idx] = 1. # nf x nnjoints #
+  minn_topk_mask[:, -5: -3] = 1.
+  basepts_idx_range = torch.arange(nn_base_pts).unsqueeze(0).unsqueeze(0)
+  minn_dist_mask = basepts_idx_range == minn_dist_idx.unsqueeze(-1) # nf x nnjoints x nnbasepts
+  # for seq 101
+  # minn_dist_mask[31:, -5, :] = minn_dist_mask[30: 31, -5, :]
+  minn_dist_mask = minn_dist_mask.float()
   
-  # ## tot base pts 
-  # tot_base_pts_trans_disp = torch.sum(
-  #   (tot_base_pts_trans[1:, :, :] - tot_base_pts_trans[:-1, :, :]) ** 2, dim=-1 # (nf - 1) x nn_base_pts displacement 
-  # )
-  # ### tot base pts trans disp ###
-  # tot_base_pts_trans_disp = torch.sqrt(tot_base_pts_trans_disp).mean(dim=-1) # (nf - 1)
-  # # tot_base_pts_trans_disp_mov_thres = 1e-20
-  # tot_base_pts_trans_disp_mov_thres = 3e-4
-  # tot_base_pts_trans_disp_mask = tot_base_pts_trans_disp >= tot_base_pts_trans_disp_mov_thres
-  # tot_base_pts_trans_disp_mask = torch.cat(
-  #   [tot_base_pts_trans_disp_mask, tot_base_pts_trans_disp_mask[-1:]], dim=0
-  # )
+  ## tot base pts 
+  tot_base_pts_trans_disp = torch.sum(
+    (tot_base_pts_trans[1:, :, :] - tot_base_pts_trans[:-1, :, :]) ** 2, dim=-1 # (nf - 1) x nn_base_pts displacement 
+  )
+  ### tot base pts trans disp ###
+  tot_base_pts_trans_disp = torch.sqrt(tot_base_pts_trans_disp).mean(dim=-1) # (nf - 1)
+  # tot_base_pts_trans_disp_mov_thres = 1e-20
+  tot_base_pts_trans_disp_mov_thres = 3e-4
+  tot_base_pts_trans_disp_mask = tot_base_pts_trans_disp >= tot_base_pts_trans_disp_mov_thres
+  tot_base_pts_trans_disp_mask = torch.cat(
+    [tot_base_pts_trans_disp_mask, tot_base_pts_trans_disp_mask[-1:]], dim=0
+  )
   
-  # attraction_mask_new = (tot_base_pts_trans_disp_mask.float().unsqueeze(-1).unsqueeze(-1) + minn_dist_mask.float()) > 1.5
+  attraction_mask_new = (tot_base_pts_trans_disp_mask.float().unsqueeze(-1).unsqueeze(-1) + minn_dist_mask.float()) > 1.5
   
   
   
-  # minn_topk_mask = (minn_dist_mask + minn_topk_mask.float().unsqueeze(-1)) > 1.5
-  # print(f"minn_dist_mask: {minn_dist_mask.size()}")
+  minn_topk_mask = (minn_dist_mask + minn_topk_mask.float().unsqueeze(-1)) > 1.5
+  print(f"minn_dist_mask: {minn_dist_mask.size()}")
   s = 1.0
   # affinity_scores = get_affinity_fr_dist(dist_joints_to_base_pts, s=s)
 
@@ -231,21 +231,21 @@ def get_optimized_hand_fr_joints_v4_anchors(joints, base_pts, tot_base_pts_trans
       print('\tRotation Smoothness Loss: {}'.format(joints_pred_loss.item()))
   
   # 
-  # print(tot_base_pts_trans.size())
-  # diff_base_pts_trans = torch.sum((tot_base_pts_trans[1:, :, :] - tot_base_pts_trans[:-1, :, :]) ** 2, dim=-1) # (nf - 1) x nn_base_pts
-  # print(f"diff_base_pts_trans: {diff_base_pts_trans.size()}")
-  # diff_base_pts_trans = diff_base_pts_trans.mean(dim=-1)
-  # diff_base_pts_trans_threshold = 1e-20
-  # diff_base_pts_trans_mask = diff_base_pts_trans > diff_base_pts_trans_threshold # (nf - 1) ### the mask of the tranformed base pts
-  # diff_base_pts_trans_mask = diff_base_pts_trans_mask.float()
-  # print(f"diff_base_pts_trans_mask: {diff_base_pts_trans_mask.size()}, diff_base_pts_trans: {diff_base_pts_trans.size()}")
-  # diff_last_frame_mask = torch.tensor([0,], dtype=torch.float32).to(diff_base_pts_trans_mask.device) + diff_base_pts_trans_mask[-1]
-  # diff_base_pts_trans_mask = torch.cat(
-  #   [diff_base_pts_trans_mask, diff_last_frame_mask], dim=0 # nf tensor
-  # )
+  print(tot_base_pts_trans.size())
+  diff_base_pts_trans = torch.sum((tot_base_pts_trans[1:, :, :] - tot_base_pts_trans[:-1, :, :]) ** 2, dim=-1) # (nf - 1) x nn_base_pts
+  print(f"diff_base_pts_trans: {diff_base_pts_trans.size()}")
+  diff_base_pts_trans = diff_base_pts_trans.mean(dim=-1)
+  diff_base_pts_trans_threshold = 1e-20
+  diff_base_pts_trans_mask = diff_base_pts_trans > diff_base_pts_trans_threshold # (nf - 1) ### the mask of the tranformed base pts
+  diff_base_pts_trans_mask = diff_base_pts_trans_mask.float()
+  print(f"diff_base_pts_trans_mask: {diff_base_pts_trans_mask.size()}, diff_base_pts_trans: {diff_base_pts_trans.size()}")
+  diff_last_frame_mask = torch.tensor([0,], dtype=torch.float32).to(diff_base_pts_trans_mask.device) + diff_base_pts_trans_mask[-1]
+  diff_base_pts_trans_mask = torch.cat(
+    [diff_base_pts_trans_mask, diff_last_frame_mask], dim=0 # nf tensor
+  )
   # attraction_mask = (diff_base_pts_trans_mask.unsqueeze(-1).unsqueeze(-1) + minn_topk_mask.float()) > 1.5
-  # attraction_mask = minn_topk_mask.float()
-  # attraction_mask = attraction_mask.float()
+  attraction_mask = minn_topk_mask.float()
+  attraction_mask = attraction_mask.float()
   
   # the direction of the normal vector and the moving direction of the object point -> whether the point should be selected
   # the contact maps of the object should be like? #
@@ -1121,7 +1121,7 @@ def reconstruct_from_file(single_seq_path):
   if st_idxes[-1] + num_cleaning_frames < nn_frames:
       st_idxes.append(nn_frames - num_cleaning_frames)
       
-  # st_idxes = [st_idxes[0]]
+  st_idxes = [st_idxes[0]]
   print(f"st_idxes: {st_idxes}")
   
   
@@ -1184,12 +1184,7 @@ def reconstruct_from_file(single_seq_path):
                   tot_data[cur_k].append(cur_data[cur_k][ :clip_ending_idxes[i_tag]])
     
     for cur_k in tot_data:
-      print(f"cur_k: {cur_k}")
-      for aa in tot_data[cur_k]:
-        print(aa.shape)
-      if cur_k in ['tot_base_pts', 'tot_base_normals']:
-        continue
-      elif cur_k in ["tot_base_pts", "tot_base_normals", "tot_obj_rot", "tot_obj_transl", "tot_obj_pcs", "tot_rhand_joints", "tot_gt_rhand_joints"]:
+      if cur_k in ["tot_base_pts", "tot_base_normals", "tot_obj_rot", "tot_obj_transl", "tot_obj_pcs", "tot_rhand_joints", "tot_gt_rhand_joints"]:
           tot_data[cur_k] = np.concatenate(tot_data[cur_k], axis=1)
       else:
           tot_data[cur_k] = np.concatenate(tot_data[cur_k], axis=0)
@@ -1207,8 +1202,8 @@ def reconstruct_from_file(single_seq_path):
     
     targets = data['targets'] # # targets # #
     outputs = data['outputs'] #  
-    # tot_base_pts = data["tot_base_pts"][0] # total base pts, total base normals #
-    # tot_base_normals = data['tot_base_normals'][0] # nn_base_normals #
+    tot_base_pts = data["tot_base_pts"][0] # total base pts, total base normals #
+    tot_base_normals = data['tot_base_normals'][0] # nn_base_normals #
     
     
     
@@ -1218,22 +1213,22 @@ def reconstruct_from_file(single_seq_path):
     tot_obj_transl = data['tot_obj_transl'][0]
     print(f"tot_obj_rot: {tot_obj_rot.shape}, tot_obj_transl: {tot_obj_transl.shape}")
     
-    # if len(tot_base_pts.shape) == 2:
-    #   # numpy array # # tot base pts #
-    #   tot_base_pts_trans = np.matmul(tot_base_pts.reshape(1, tot_base_pts.shape[0], 3), tot_obj_rot) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])
-    #   tot_base_pts = np.matmul(tot_base_pts, tot_obj_rot[0]) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])[0] 
+    if len(tot_base_pts.shape) == 2:
+      # numpy array # # tot base pts #
+      tot_base_pts_trans = np.matmul(tot_base_pts.reshape(1, tot_base_pts.shape[0], 3), tot_obj_rot) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])
+      tot_base_pts = np.matmul(tot_base_pts, tot_obj_rot[0]) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])[0] 
       
-    #   tot_base_normals_trans = np.matmul( # # 
-    #     tot_base_normals.reshape(1, tot_base_normals.shape[0], 3), tot_obj_rot
-    #   ) 
-    # else:
-    #   print(f"tot_base_pts: {tot_base_pts.shape}, tot_obj_rot: {tot_obj_rot.shape}, tot_obj_transl: {tot_obj_transl.shape}")
-    #   tot_base_pts_trans = np.matmul(tot_base_pts, tot_obj_rot) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])
-    #   tot_base_pts = np.matmul(tot_base_pts, tot_obj_rot) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])
+      tot_base_normals_trans = np.matmul( # # 
+        tot_base_normals.reshape(1, tot_base_normals.shape[0], 3), tot_obj_rot
+      ) 
+    else:
+      print(f"tot_base_pts: {tot_base_pts.shape}, tot_obj_rot: {tot_obj_rot.shape}, tot_obj_transl: {tot_obj_transl.shape}")
+      tot_base_pts_trans = np.matmul(tot_base_pts, tot_obj_rot) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])
+      tot_base_pts = np.matmul(tot_base_pts, tot_obj_rot) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1])
       
-    #   tot_base_normals_trans = np.matmul(
-    #     tot_base_normals, tot_obj_rot
-    #   )
+      tot_base_normals_trans = np.matmul(
+        tot_base_normals, tot_obj_rot
+      )
 
 
     
@@ -1242,7 +1237,7 @@ def reconstruct_from_file(single_seq_path):
     
     targets = np.matmul(targets, tot_obj_rot) + tot_obj_transl.reshape(tot_obj_transl.shape[0], 1, tot_obj_transl.shape[1]) # ws x nn_verts x 3 #
     # denoise relative positions 
-    # print(f"tot_base_pts: {tot_base_pts.shape}")
+    print(f"tot_base_pts: {tot_base_pts.shape}")
     
     
     #### obj_verts_trans, obj_faces ####
@@ -1269,7 +1264,7 @@ def reconstruct_from_file(single_seq_path):
     with_contact_opt = True
     with_ctx_mask = False
 
-    bf_ct_optimized_dict, bf_proj_optimized_dict, optimized_dict = get_optimized_hand_fr_joints_v4_anchors(outputs, None, None, None, with_contact_opt=with_contact_opt, nn_hand_params=nn_hand_params, rt_vars=True, with_proj=with_proj, obj_verts_trans=obj_verts_trans, obj_faces=obj_faces, with_params_smoothing=with_params_smoothing, dist_thres=dist_thres, with_ctx_mask=with_ctx_mask)
+    bf_ct_optimized_dict, bf_proj_optimized_dict, optimized_dict = get_optimized_hand_fr_joints_v4_anchors(outputs, tot_base_pts, tot_base_pts_trans, tot_base_normals_trans, with_contact_opt=with_contact_opt, nn_hand_params=nn_hand_params, rt_vars=True, with_proj=with_proj, obj_verts_trans=obj_verts_trans, obj_faces=obj_faces, with_params_smoothing=with_params_smoothing, dist_thres=dist_thres, with_ctx_mask=with_ctx_mask)
 
     
 
@@ -1279,12 +1274,12 @@ def reconstruct_from_file(single_seq_path):
     optimized_sv_infos.update(bf_ct_optimized_dict)
     optimized_sv_infos.update(bf_proj_optimized_dict)
     optimized_sv_infos.update(optimized_dict)
-    # optimized_sv_infos.update(
-    #   {
-    #     'tot_base_pts_trans': tot_base_pts_trans,
-    #     'tot_base_normals_trans': tot_base_normals_trans
-    #   }
-    # )
+    optimized_sv_infos.update(
+      {
+        'tot_base_pts_trans': tot_base_pts_trans,
+        'tot_base_normals_trans': tot_base_normals_trans
+      }
+    )
     
     
     optimized_sv_infos.update({'predicted_info': data})