expression friendly method for driving single images

liveportrait/live_portrait_pipeline.py

@@ -14,6 +14,7 @@ from .utils.camera import get_rotation_matrix
 from .live_portrait_wrapper import LivePortraitWrapper
 from .utils.retargeting_utils import calc_eye_close_ratio, calc_lip_close_ratio
 from .utils.filter import smooth
+from .utils.helper import calc_motion_multiplier
 
 import os
 script_directory = os.path.dirname(os.path.abspath(__file__))
@@ -38,7 +39,7 @@ class LivePortraitPipeline(object):
         )
 
     def execute(
-        self, driving_images, crop_info, driving_landmarks, delta_multiplier, relative_motion_mode, driving_smooth_observation_variance, mismatch_method="constant", 
+        self, driving_images, crop_info, driving_landmarks, delta_multiplier, relative_motion_mode, driving_smooth_observation_variance, mismatch_method="constant", expression_friendly=False, driving_multiplier=1.0, 
     ):
         inference_cfg = self.live_portrait_wrapper.cfg
         device = inference_cfg.device_id
@@ -174,6 +175,15 @@ class LivePortraitPipeline(object):
             delta_new = delta_new * delta_multiplier
             
             x_d_i_new = scale_new * (x_c_s @ R_new + delta_new) + t_new
+
+            if expression_friendly:
+                if i == 0:
+                    x_d_0_new = x_d_i_new
+                    motion_multiplier = calc_motion_multiplier(x_s, x_d_0_new)
+                    motion_multiplier *= driving_multiplier
+                x_d_diff = (x_d_i_new - x_d_0_new) * motion_multiplier
+                x_d_i_new = x_d_diff + x_s
+
             if (
                 not inference_cfg.flag_stitching
                 and not inference_cfg.flag_eye_retargeting

liveportrait/utils/helper.py

@@ -4,10 +4,12 @@
 utility functions and classes to handle feature extraction and model loading
 """
 
-import os.path as osp
 import cv2
 import torch
+import numpy as np
+from typing import Union
 from collections import OrderedDict
+from scipy.spatial import ConvexHull # pylint: disable=E0401,E0611
 
 def squeeze_tensor_to_numpy(tensor):
     out = tensor.data.squeeze(0).cpu().numpy()
@@ -71,3 +73,24 @@ def resize_to_limit(img, max_dim=1280, n=2):
     if new_h != img.shape[0] or new_w != img.shape[1]:
         img = img[:new_h, :new_w]
     return img
+
+def tensor_to_numpy(data: Union[np.ndarray, torch.Tensor]) -> np.ndarray:
+    """transform torch.Tensor into numpy.ndarray"""
+    if isinstance(data, torch.Tensor):
+        return data.data.cpu().numpy()
+    return data
+
+def calc_motion_multiplier(
+    kp_source: Union[np.ndarray, torch.Tensor],
+    kp_driving_initial: Union[np.ndarray, torch.Tensor]
+) -> float:
+    """calculate motion_multiplier based on the source image and the first driving frame"""
+    kp_source_np = tensor_to_numpy(kp_source)
+    kp_driving_initial_np = tensor_to_numpy(kp_driving_initial)
+
+    source_area = ConvexHull(kp_source_np.squeeze(0)).volume
+    driving_area = ConvexHull(kp_driving_initial_np.squeeze(0)).volume
+    motion_multiplier = np.sqrt(source_area) / np.sqrt(driving_area)
+    # motion_multiplier = np.cbrt(source_area) / np.cbrt(driving_area)
+
+    return motion_multiplier

nodes.py

@@ -280,6 +280,8 @@ class LivePortraitProcess:
             
             "optional": {
                 "opt_retargeting_info": ("RETARGETINGINFO", {"default": None}),
+                "expression_friendly": ("BOOLEAN", {"default": False}),
+                "expression_friendly_multiplier": ("FLOAT", {"default": 1.0, "min": 0.01, "max": 100.0, "step": 0.001}),
             }
         }
 
@@ -308,9 +310,13 @@ class LivePortraitProcess:
         delta_multiplier: float = 1.0,
         mismatch_method: str = "constant",
         opt_retargeting_info: dict = None,
+        expression_friendly: bool = False,
+        expression_friendly_multiplier: float = 1.0,
     ):
         if driving_images.shape[0] < source_image.shape[0]:
             raise ValueError("The number of driving images should be larger than the number of source images.")
+        if expression_friendly and source_image.shape[0] > 1:
+            raise ValueError("expression_friendly works only with single source image")
         
         if opt_retargeting_info is not None:
             pipeline.live_portrait_wrapper.cfg.flag_eye_retargeting = opt_retargeting_info["eye_retargeting"]
@@ -347,7 +353,9 @@ class LivePortraitProcess:
             delta_multiplier,
             relative_motion_mode,
             driving_smooth_observation_variance,
-            mismatch_method
+            mismatch_method,
+            expression_friendly=expression_friendly,
+            driving_multiplier=expression_friendly_multiplier,
         )
 
         total_frames = len(out["out_list"])