We provide Pytorch
implementation for sRender: rendering facial sketches
Citation:
Meimei Shang, Fei Gao *, Xiang Li, Jingjie Zhu, Lingna Dai. Bridging Unpaired Facial Photos And Sketches By Line-drawings. 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 6-11 June 2021, Toronto, Ontario, Canada. (Accepted)
[paper@arXiv] [project@github]
left: croquis style; right: charcoal style (a)Sketches ; (b)Generated Line-drawings ; (c)Synthesized Sketches
(a)Oilpaintings ; (b)Generated Line-drawings ; (c)Synthesized Oil Paintings
left: croquis style; right: charcoal style (a)Photos ; (b)Generated Line-drawings ; (c)Synthesized Sketches
(a)Photos ; (b)Generated Line-drawings ; (c)Synthesized Oil Paintings
- Croquis sketches generated by our sRender and unpaired I2I translation methods
charcoal_style
for systhesising charcoal style images,it contains sRender w/o Lstr for model without stroke_loss and sRender for model with stroke_loss correspodinglycroquis_style
for systhesising croquis style images,it contains sRenderPix2Pix, sRender w/o Lstr and sRender- download dataset gray for charcol_style, binary for croquis_style
- download pretrain model and put them in ./checkpoints for test
- for model with stroke_loss, you should download stroke_model and specify model root in ./models/pix2pixHD_model for net_c.load_state_dict
- you can modify options/base_option to specify --dataroot, then run train.py or test.py
- Gray line-drawings for charcol_style from GooleDrive
- Binary line-drawings for croquis_style from GooleDrive
- charcoal_style from GooleDrive
- croquis_style from GooleDrive
- oil_painting_style from GooleDrive
- Our synthesis result for croquis and charcoal style can be downloaded from Goole Drive
- Our synthesis result for oil painting style can be downloaded from Goole Drive
- Our code is inspired by the NVIDIA/pix2pixHD repository.