|
@@ -23,7 +23,7 @@ def encode_segmentation_rgb(segmentation, no_neck=True):
|
|
mouth_map[valid_index] = 255
|
|
mouth_map[valid_index] = 255
|
|
# valid_index = np.where(parse==hair_id)
|
|
# valid_index = np.where(parse==hair_id)
|
|
# hair_map[valid_index] = 255
|
|
# hair_map[valid_index] = 255
|
|
-
|
|
|
|
|
|
+ #return np.stack([face_map, mouth_map,hair_map], axis=2)
|
|
return np.stack([face_map, mouth_map], axis=2)
|
|
return np.stack([face_map, mouth_map], axis=2)
|
|
|
|
|
|
|
|
|
|
@@ -108,7 +108,7 @@ def reverse2wholeimage(b_align_crop_tenor_list,swaped_imgs, mats, crop_size, ori
|
|
parsing = out.squeeze(0).detach().cpu().numpy().argmax(0)
|
|
parsing = out.squeeze(0).detach().cpu().numpy().argmax(0)
|
|
vis_parsing_anno = parsing.copy().astype(np.uint8)
|
|
vis_parsing_anno = parsing.copy().astype(np.uint8)
|
|
tgt_mask = encode_segmentation_rgb(vis_parsing_anno)
|
|
tgt_mask = encode_segmentation_rgb(vis_parsing_anno)
|
|
- if tgt_mask.sum() != 0:
|
|
|
|
|
|
+ if tgt_mask.sum() >= 5000:
|
|
# face_mask_tensor = tgt_mask[...,0] + tgt_mask[...,1]
|
|
# face_mask_tensor = tgt_mask[...,0] + tgt_mask[...,1]
|
|
target_mask = cv2.resize(tgt_mask, (224, 224))
|
|
target_mask = cv2.resize(tgt_mask, (224, 224))
|
|
# print(source_img)
|
|
# print(source_img)
|