How to use my own pictures for 3D reconstruction

[jly0810]

  I want to use my photos to generate a 3D portrait model. I learned that in addition to photos, I also need several other files to input. 
  I have used the method here [https://github.com/sicxu/Deep3dPortrait/issues/9](url) to generate landmark.txt file.But I don't know how to generate the detection.txt files and the .mat file.
 I am a novice. I hope you can give me as detailed steps as possible to answer my doubts. Thank you very much for your help.

[Ravindu-Yasas-Nagasinghe]

Have you found a way to create detection.txt file? @jly0810 If so please mention the method.

[zwyking]

Hi, I have the same question! Have you find the way to generate them (.mat, landmark.txt)?

[Bai426]

@sicxu Hi, Dr. Xu, this is a great work. I followed #25 and LeoGitGuy's suggestion, but the depth map has problem.

/home/neural_image/Deep3dPortrait/utils/preprocess.py:27: FutureWarning: rcond parameter will change to the default of machine precision times max(M, N) where M and N are the input matrix dimensions. To use the future default and silence this warning we advise to pass rcond=None, to keep using the old, explicitly pass rcond=-1. k, , , _ = np.linalg.lstsq(A, b) (239, 239, 3) (159, 159, 3) (79, 79) Traceback (most recent call last): File "step1_recon_3d_face.py", line 126, in face_recon(input_path, save_path, vis_path) File "step1_recon_3d_face.py", line 74, in face_recon crop_img, crop_mask, inv_params, crop_lm, crop_param = headrecon_preprocess_withmask(img, mask, landmarks_2d, center2d.reshape([2]), s*s_factor/100) File "/home/neural_image/Deep3dPortrait/utils/preprocess.py", line 171, in headrecon_preprocess_withmask mask, lm.copy(), head_center.reshape([2]), scale, crp_w, crp_h, interpolation=cv2.INTER_NEAREST) File "/home/neural_image/Deep3dPortrait/utils/preprocess.py", line 69, in align_img cropped_img, _ = crop(img, bbox) File "/home/neural_image/Deep3dPortrait/utils/preprocess.py", line 53, in crop padded_img, padded_bbox = img_padding(img, bbox) File "/home/neural_image/Deep3dPortrait/utils/preprocess.py", line 49, in img_padding assert(bbox[0] >= 0 and bbox[1] >= 0) AssertionError_

Can you give me some advice? Thank you very much!

Best, Mike

[syysummer]

Hi, I have the same question! Have you find the way to generate them (.mat, landmark.txt)?

Have you found the method to generate landmark.txt, detection.txt, and .mat?

[Bai426]

Just follow https://github.com/sicxu/Deep3dPortrait/issues/25 and other issues in this repo, you can generate these files.

[syysummer]

Just follow #25 and other issues in this repo, you can generate these files.

Thank you very much for your answer! I have obtained the coordinates of 68 key points, but when I obtained the initial .mat file, I found that "https://arxiv.org/pdf/1906.01342.pdf" was inaccessible. Do you have any ideas?