No such file or directory: '../test_data/2/samples/2d_pose_result_hrnet.json'

[Rietchie0119]

Hello, when I run

python preprocess_img_for_inversion.py --test_data_dir=$PROJECT_ROOT$/test_data

I get the following errors:

Traceback (most recent call last):
  File "tools/get_keypoints.py", line 264, in <module>
    main()
  File "tools/get_keypoints.py", line 248, in main
    with open(json_save_path, 'w') as f:
FileNotFoundError: [Errno 2] No such file or directory: '../test_data/2/samples/2d_pose_result_hrnet.json'

Looks like I'm missing json file but I followed everything as instruction says. How can I solve this problem?

[oneThousand1000]

Hi, it seems there something wrong with your 2d pose extraction so that the json file is not generated.

Could you please do the following steps?

1. Delete everything in the test/2 dir, except the prompt.txt
2. Rerun the random image generation step and the preprocess_img_for_inversion step.
3. Show me all errors you meet (especially the first one)

[Rietchie0119]

Thanks for your response. I reran the code but I am still getting the same error. Does the following command supposed to create "../test_data/0/samples/2d_pose_result_hrnet.json" file?

python get_test_data_df.py --test_data_dir ../test_data --sample_num 6  --scale 5 --df_ckpt $PATH_of_Realistic_Vision_V5.1_noVAE$/realisticVisionV51_v51VAE.ckpt

By the way I noticed that I was using files from "SMPL_python_v.1.1.0.zip" and not from "SMPL_python_v.1.0.0.zip" I changed the files to correct ones but I still get the same error. Following is the full text when I run

python preprocess_img_for_inversion.py --test_data_dir=../test_data

/home/jeonguk/Portrait3D/data_processing
/home/jeonguk/Portrait3D/data_processing
python tools/get_keypoints.py  --cfg experiments/coco/higher_hrnet/w32_512_adam_lr1e-3.yaml  --input_dir ../test_data/0/samples TEST.MODEL_FILE models/pytorch/pose_coco/pose_higher_hrnet_w32_512.pth 
=> creating output/coco_kpt/pose_higher_hrnet/w32_512_adam_lr1e-3
=> creating log/coco_kpt/pose_higher_hrnet/w32_512_adam_lr1e-3_2024-06-01-08-28
Namespace(cfg='experiments/coco/higher_hrnet/w32_512_adam_lr1e-3.yaml', input_dir='../test_data/0/samples', opts=['TEST.MODEL_FILE', 'models/pytorch/pose_coco/pose_higher_hrnet_w32_512.pth'])
=> loading model from models/pytorch/pose_coco/pose_higher_hrnet_w32_512.pth
json_save_path ../test_data/0/samples/2d_pose_result_hrnet.json
meta_save_path ../test_data/0/samples/meta_data.json
0it [00:00, ?it/s]
Traceback (most recent call last):
  File "tools/get_keypoints.py", line 264, in <module>
    main()
  File "tools/get_keypoints.py", line 248, in main
    with open(json_save_path, 'w') as f:
FileNotFoundError: [Errno 2] No such file or directory: '../test_data/0/samples/2d_pose_result_hrnet.json'
python detect_head_bbox.py --weights crowdhuman_yolov5m.pt --source ../test_data/0/samples   --heads
Namespace(agnostic_nms=False, augment=False, classes=None, conf_thres=0.25, device='', exist_ok=False, heads=True, img_size=640, iou_thres=0.45, name='exp', person=False, project='runs/detect', save_conf=False, save_txt=False, source='../test_data/0/samples', update=False, view_img=False, weights=['crowdhuman_yolov5m.pt'])
Traceback (most recent call last):
  File "detect_head_bbox.py", line 294, in <module>
    detect()
  File "detect_head_bbox.py", line 26, in detect
    with open(skeleton_path) as f:
FileNotFoundError: [Errno 2] No such file or directory: '../test_data/0/samples/2d_pose_result_hrnet.json'
/home/jeonguk/Portrait3D/data_processing
Traceback (most recent call last):
  File "runmy.py", line 19, in <module>
    raise Exception('path or head_box_path not exists')
Exception: path or head_box_path not exists
/home/jeonguk/Portrait3D/data_processing
python new_crop_use_densepose.py --gpu 0  --input_dir  ../test_data/0/samples  --output_dir  ../test_data/0/samples_new_crop  --data_dir /home/jeonguk/Portrait3D/data_processing/data
output dir:  /home/jeonguk/Portrait3D/data_processing/main/../output/exp_06-01_08-29
exp_dir:  
>>> Using GPU: 0
/home/jeonguk/Portrait3D/data_processing/main/../common/utils/smplpytorch/smplpytorch/pytorch/smpl_layer.py:41: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at  ../torch/csrc/utils/tensor_numpy.cpp:189.)
  torch.Tensor(smpl_data['betas'].r).unsqueeze(0))
>>>>>>> output_dir ../test_data/0/samples_new_crop
Traceback (most recent call last):
  File "new_crop_use_densepose.py", line 170, in <module>
    with open(result_json_path, 'r') as f:
FileNotFoundError: [Errno 2] No such file or directory: '../test_data/0/samples/result.json'
0it [00:00, ?it/s]

[Rietchie0119]

Ok I found the problem. I passed relative path for the following command and it did not work.

python preprocess_img_for_inversion.py --test_data_dir=$PROJECT_ROOT$/test_data

After passing absolute path, it worked totally fine!