How to obtain landmark_embedding.npy?

[StarSapph1re]

I attempted to run the train_splatting_avatar.py but it seems a landmark_embedding file is needed. creating the FLAME Decoder Traceback (most recent call last): File "/root/SplattingAvatar/train_splatting_avatar.py", line 43, in frameset_train = make_frameset_data(config.dataset, split='train') File "/root/SplattingAvatar/dataset/dataset_helper.py", line 20, in make_frameset_data frameset = IMavatarDataset(dataset_config, split=split, frm_list=frm_list) File "/root/SplattingAvatar/dataset/imavatar_data.py", line 65, in init self.load_flame_json() File "/root/SplattingAvatar/dataset/imavatar_data.py", line 111, in load_flame_json self.flame = FLAME('model/imavatar/FLAME2020/generic_model.pkl', File "/root/SplattingAvatar/model/imavatar/flame.py", line 51, in init lmk_embeddings = np.load(lmk_embedding_path, allow_pickle=True, encoding='latin1') File "/root/miniconda3/envs/splatting/lib/python3.9/site-packages/numpy/lib/npyio.py", line 390, in load fid = stack.enter_context(open(os_fspath(file), "rb")) FileNotFoundError: [Errno 2] No such file or directory: 'model/imavatar/FLAME2020/landmark_embedding.npy' There's no FLAME2020 in the original code and I created it, copying generic_model.pkl into it, but how to obtain landmark_embedding.npy? Thanks so much for your greating work. Looking forward to your reply.

[Loppas]

I ran into the same issue. You can grab it here:

https://github.com/yfeng95/DECA/blob/master/data/landmark_embedding.npy

[StarSapph1re]

I ran into the same issue. You can grab it here:

https://github.com/yfeng95/DECA/blob/master/data/landmark_embedding.npy

Thank you so much:)