Improved risen hand detection and fixes

monoloco/activity.py

@@ -66,22 +66,51 @@ def social_interactions(idx, centers, angles, dds, stds=None, social_distance=Fa
     return False
 
 
-def is_raising_hand(keypoint):
+def is_raising_hand(kp):
     """
     Returns flag of alert if someone raises their hand
     """
+    x=0
+    y=1
+
+    nose = 0
+    l_ear = 3
     l_shoulder = 5
+    l_elbow = 7
     l_hand = 9
+    r_ear = 4
     r_shoulder = 6
+    r_elbow = 8
     r_hand = 10
 
-    if keypoint[1][l_hand] < keypoint[1][l_shoulder] and keypoint[1][r_hand] < keypoint[1][r_shoulder]:
+    head_width = kp[x][l_ear]- kp[x][r_ear]
+    head_top = (4/5) * (kp[y][nose] - head_width)
+    
+    l_forearm = [kp[x][l_hand] - kp[x][l_elbow], kp[y][l_hand] - kp[y][l_elbow]]
+    l_arm = [kp[x][l_shoulder] - kp[x][l_elbow], kp[y][l_shoulder] - kp[y][l_elbow]]     
+
+    r_forearm = [kp[x][r_hand] - kp[x][r_elbow], kp[y][r_hand] - kp[y][r_elbow]]
+    r_arm = [kp[x][r_shoulder] - kp[x][r_elbow], kp[y][r_shoulder] - kp[y][r_elbow]]     
+
+    l_angle = (90/np.pi) * np.arccos(np.dot(l_forearm/np.linalg.norm(l_forearm), l_arm/np.linalg.norm(l_arm)))
+    r_angle = (90/np.pi) * np.arccos(np.dot(r_forearm/np.linalg.norm(r_forearm), r_arm/np.linalg.norm(r_arm)))
+
+    is_l_up = kp[y][l_hand] < kp[y][l_shoulder] 
+    is_r_up = kp[y][r_hand] < kp[y][r_shoulder] 
+    
+    l_too_close = kp[x][l_hand] <= kp[x][l_shoulder] and kp[y][l_hand]>=head_top
+    r_too_close = kp[x][r_hand] >= kp[x][r_shoulder] and kp[y][r_hand]>=head_top
+
+    is_left_risen = is_l_up and l_angle >= 30 and not l_too_close 
+    is_right_risen = is_r_up and r_angle >= 30 and not r_too_close
+
+    if is_left_risen and is_right_risen:
         return 'both'
 
-    if keypoint[1][l_hand] < keypoint[1][l_shoulder]:
+    if is_left_risen:
         return 'left'
 
-    if keypoint[1][r_hand] < keypoint[1][r_shoulder]:
+    if is_right_risen:
         return 'right'
 
     return 'none'

monoloco/run.py

@@ -19,7 +19,7 @@ def cli():
     predict_parser.add_argument('images', nargs='*', help='input images')
     predict_parser.add_argument('--glob', help='glob expression for input images (for many images)')
     predict_parser.add_argument('-o', '--output-directory', help='Output directory')
-    predict_parser.add_argument('--output_types', nargs='+', default=['json'],
+    predict_parser.add_argument('--output_types', nargs='+',
                                 help='what to output: json keypoints skeleton for Pifpaf'
                                      'json bird front or multi for MonStereo')
     predict_parser.add_argument('--no_save', help='to show images', action='store_true')
@@ -131,6 +131,8 @@ def main():
             from .visuals.webcam import webcam
             webcam(args)
         else:
+            if args.output_types is None:
+                args.output_types = ['json']
             from .predict import predict
             predict(args)
 

monoloco/visuals/pifpaf_show.py

@@ -103,7 +103,7 @@ class KeypointPainter(object):
                     linewidth = np.sqrt((x[9]-x[7])**2 + (y[9]-y[7])**2)
                 if ((connection[0] == 6 and connection[1] == 8) or (connection[0] == 8 and connection[1] == 10)) and raise_hand in ['right', 'both']:
                     c = 'yellow'
-                    linewidth = np.sqrt((x[9]-x[7])**2 + (y[9]-y[7])**2)
+                    linewidth = np.sqrt((x[10]-x[8])**2 + (y[10]-y[8])**2)
                 if self.color_connections:
                     c = matplotlib.cm.get_cmap('tab20')(ci / len(self.skeleton))
                 if np.all(v[connection] > self.dashed_threshold):
@@ -153,7 +153,7 @@ class KeypointPainter(object):
             ax.text(x1, y1, '{:.4f}'.format(score), fontsize=8, color=color)
 
     @staticmethod
-    def _draw_text(ax, x, y, v, text, color):
+    def _draw_text(ax, x, y, v, text, color, fontsize=8):
         if not np.any(v > 0):
             return
 
@@ -167,7 +167,7 @@ class KeypointPainter(object):
             y1 -= 2.0
             y2 += 2.0
 
-        ax.text(x1 + 2, y1 - 2, text, fontsize=8,
+        ax.text(x1 + 2, y1 - 2, text, fontsize=fontsize,
                 color='white', bbox={'facecolor': color, 'alpha': 0.5, 'linewidth': 0})
 
     @staticmethod
@@ -208,7 +208,7 @@ class KeypointPainter(object):
             if score is not None:
                 z_str = str(score).split(sep='.')
                 text = z_str[0] + '.' + z_str[1][0]
-                self._draw_text(ax, x-2, y, v, text, color)
+                self._draw_text(ax, x[1:3], y[1:3]-5, v[1:3], text, color, fontsize=16)
             if self.show_box:
                 score = scores[i] if scores is not None else None
                 self._draw_box(ax, x, y, v, color, score)

monoloco/visuals/printer.py

@@ -417,6 +417,7 @@ class Printer:
             uv_max = [0., float(self.height)]
             xyz_max = pixel_to_camera(uv_max, self.kk, self.z_max)
             x_max = abs(xyz_max[0])  # shortcut to avoid oval circles in case of different kk
+            x_max=6
             corr = round(float(x_max / 3))
             ax.plot([0, x_max], [0, self.z_max], 'k--')
             ax.plot([0, -x_max], [0, self.z_max], 'k--')

monoloco/visuals/webcam.py

@@ -13,7 +13,10 @@ import logging
 import torch
 import matplotlib.pyplot as plt
 from PIL import Image
-import cv2
+try:
+    import cv2
+except ImportError:
+    cv2 = None
 
 from openpifpaf import decoder, network, visualizer, show, logger
 import openpifpaf.datasets as datasets
@@ -34,6 +37,15 @@ def factory_from_args(args):
 
     logger.configure(args, LOG)  # logger first
     
+    if args.output_types is None:
+        args.output_types = ['multi']
+
+    assert 'bird' not in args.output_types
+    if 'json' not in args.output_types:
+        assert len(args.output_types) is 1
+    else:
+        assert len(args.output_types) < 3
+
     # Devices
     args.device = torch.device('cpu')
     args.pin_memory = False
@@ -67,6 +79,7 @@ def factory_from_args(args):
 def webcam(args):
     
     assert args.mode in ('mono')
+
     args, dic_models = factory_from_args(args)
 
     # Load Models
@@ -175,7 +188,7 @@ class Visualizer:
                 axes[1].lines = [axes[1].lines[0], axes[1].lines[1]]
                 axes[1].texts = []
 
-            if dic_out:
+            if dic_out and dic_out['dds_pred']:
                 printer._process_results(dic_out)
                 printer.draw(figures, axes, image, dic_out, pifpaf_outs['left'])
                 mypause(0.01)