Fix bug in smpl_transfer.py (fudan-generative-vision#103)

ModelAdvisor · May 1, 2024 · ad5c78f · ad5c78f
1 parent d77e274
commit ad5c78f
Showing 1 changed file with 3 additions and 2 deletions.
diff --git a/scripts/data_processors/smpl/smpl_transfer.py b/scripts/data_processors/smpl/smpl_transfer.py
@@ -99,13 +99,14 @@
             result_dict_list = np.load(group_smpl_path, allow_pickle=True)
             result_dict_first = np.load(driving_paths[0], allow_pickle=True).item()
             i = 0
-            for smpl_outs, cam_t, file_path in tqdm(
-                zip(result_dict_list["smpl"], result_dict_list["camera"], driving_paths)
+            for smpl_outs, cam_t, foc_len, file_path in tqdm(
+                zip(result_dict_list["smpl"], result_dict_list["camera"], result_dict_list["scaled_focal_length"], driving_paths)
             ):
                 img_fn, _ = os.path.splitext(os.path.basename(file_path))
                 result_dict = {key: value for key, value in result_dict_first.items()}
                 result_dict["smpls"] = smpl_outs
                 result_dict["cam_t"] = cam_t
+                result_dict["scaled_focal_length"] = foc_len
                 if args.view_transfer:
                     scaled_focal_length = reference_dict["scaled_focal_length"]
                     result_dict["cam_t"] = reference_dict["cam_t"]