Skip to content

Latest commit

 

History

History
48 lines (36 loc) · 4.32 KB

DATASET.md

File metadata and controls

48 lines (36 loc) · 4.32 KB
Raw

Pose with Style: human reposing with pose-guided StyleGAN2

Dataset and Downloads

  1. Download images:

    1. Download img_highres.zip from In-shop Clothes Retrieval Benchmark. You will need to follow the download instructions to unzip the file. Unzip file in DATASET/DeepFashion_highres/img_highres

    2. Download the train/test data: train.lst, test.lst, and fashion-pairs-test.csv. Put in DATASET/DeepFashion_highres/tools. Note: because not all training images had their densepose detected we used a slightly modified training pairs file fashion-pairs-test.csv.

    3. Split the train/test dataset using:

      python util/generate_fashion_datasets.py --dataroot DATASET/DeepFashion_highres

      This will save the train/test images in DeepFashion_highres/train and DeepFashion_highres/test.

  2. Compute densepose:

    1. Install detectron2 following their installation instructions.

    2. Use apply net from densepose and save the train/test results to a pickle file. Make sure you download densepose_rcnn_R_101_FPN_DL_s1x.pkl.

    3. Copy util/pickle2png.py into your detectron2 DensePose project directory. Using the DensePose environment, convert the pickle file to densepose png images and save the results in DATASET/densepose directory, using:

    python pickle2png.py --pickle_file train_output.pkl --save_path DATASET/densepose/train

  3. Compute human foreground mask. Save results in silhouette directory. Or you can download our computed silhouettes for the training set and testing set.

  4. Compute UV space coordinates:

    1. Compute UV space partial coordinates in the resolution 512x512.

      1. Download the UV space - 2D look up map and save it in util folder.
      2. Compute partial coordinates:
      python util/dp2coor.py --image_file DATASET/DeepFashion_highres/tools/train.lst --dp_path DATASET/densepose/train --save_path DATASET/partial_coordinates/train

    2. Complete the UV space coordinates offline, for faster training.

      1. Download the pretrained coordinate completion model from here.
      2. Complete the partial coordinates.
      python util/complete_coor.py --dataroot DATASET/DeepFashion_highres --coordinates_path DATASET/partial_coordinates --image_file DATASET/DeepFashion_highres/tools/train.lst --phase train --save_path DATASET/complete_coordinates --pretrained_model /path/to/CCM_epoch50.pt

  5. Download the following in DATASET/resources, to apply Face Identity loss:

    1. download the pre-computed required transformation (T) to align and crop the face.
    2. Download sphereface net pretrained model.

Note: we provide the DeepFashion train/test split of StylePoseGAN [Sarkar et al. 2021]: train pairs, and test pairs.