Fixed a bug that made contacts too stringent

deepof/data.py

@@ -56,7 +56,7 @@ class project:
         exclude_bodyparts: List = tuple([""]),
         exp_conditions: dict = None,
         interpolate_outliers: str = "MA",
-        interpolation_limit: int = 15,
+        interpolation_limit: int = 5,
         interpolation_std: int = 2,
         likelihood_tol: float = 0.75,
         model: str = "mouse_topview",

deepof/pose_utils.py

@@ -654,15 +654,16 @@ def tag_rulebased_frames(
             cv2.ellipse(frame, arena[0], arena[1], arena[2], 0, 360, (0, 255, 0), 3)
             # Print body parts for debuging
             for bpart in coords.columns.levels[0]:
-                cv2.circle(
-                    frame,
-                    (int(coords[bpart]["x"][fnum]), int(coords[bpart]["y"][fnum])),
-                    radius=3,
-                    color=(
-                        (255, 0, 0) if bpart.startswith(animal_ids[0]) else (0, 0, 255)
-                    ),
-                    thickness=-1,
-                )
+                if not np.isnan(coords[bpart]["x"][fnum]):
+                    cv2.circle(
+                        frame,
+                        (int(coords[bpart]["x"][fnum]), int(coords[bpart]["y"][fnum])),
+                        radius=3,
+                        color=(
+                            (255, 0, 0) if bpart.startswith(animal_ids[0]) else (0, 0, 255)
+                        ),
+                        thickness=-1,
+                    )
             # Print frame number
             write_on_frame("Frame " + str(fnum), corners["downleft"])
 

deepof/utils.py

@@ -395,9 +395,9 @@ def mask_outliers(
     mask = None
 
     if mode == "and":
-        mask = outlier_mask_x & outlier_mask_y & outlier_mask_l
+        mask = (outlier_mask_x & outlier_mask_y) | outlier_mask_l
     elif mode == "or":
-        mask = (outlier_mask_x | outlier_mask_y) & outlier_mask_l
+        mask = (outlier_mask_x | outlier_mask_y) | outlier_mask_l
 
     return mask
 
@@ -490,6 +490,7 @@ def interpolate_outliers(
     interpolated_exp.interpolate(
         method="linear", limit=limit, limit_direction="both", inplace=True
     )
+    interpolated_exp[np.isnan(interpolated_exp)] = np.nan
 
     return interpolated_exp