update torch conditional import

config/prototype_raft.json

@@ -17,7 +17,7 @@
     "nominal_ov_pos": 100,
     "n": [3,3],
     "row_tol": 0.6,
-    "box_size": [250, 250]
+    "box_size": [220, 220]
   },
   "tracker": {
     "adaptive": true,

promote_video.py

@@ -127,7 +127,7 @@ def draw_magnitude_map(vfield, threshold=3.0, colormap=cv2.COLORMAP_JET) -> np.n
 
 
 
-npoints = 100  # 图表数据长度
+npoints = 400  # 图表数据长度
 plot_threshold = 0  # only plot force values above this (below → 0)
 pg.setConfigOption('background', bg_color_hex)
 pg.setConfigOption('foreground', bg_color_hex)
@@ -154,7 +154,7 @@ def main():
     pyqt_writer = cv2.VideoWriter(args.video.replace(".mp4", "_pyqt.mp4"), cv2.VideoWriter_fourcc(*'mp4v'), fps_video, (1600, 1200))
     depth_writer = cv2.VideoWriter(args.video.replace(".mp4", "_depth.mp4"), cv2.VideoWriter_fourcc(*'mp4v'),fps_video, (384, 384))
 
-    sensor1 = orisys.Sensor(args.video, config_name="./config/prototype_of3.json", cal_path= args.video.replace(".mp4", ".npy"), isstitch=True, cuda=True, verbose=True)
+    sensor1 = orisys.Sensor(args.video, config_name="./config/prototype_raft.json", cal_path= args.video.replace(".mp4", ".npy"), isstitch=True, cuda=True, verbose=True)
 
     start_time = time.time()
     is_save = False
@@ -203,6 +203,10 @@ def main():
             data[:-1, 1, i] = data[1:, 1, i]  # shift force data left
         data[-1, 0, :] = frame_count/fps_video  # set current time for all plots
         data[-1, 1, 0] = fn
+        if abs(fx) - data[-2, 1, 1] > 100000:
+            fx = data[-2, 1, 1] + data[-2, 1, 1]*0.5
+        if abs(fy) - data[-2, 1, 2] > 100000:
+            fy = data[-2, 1, 2] + data[-2, 1, 2]*0.5
         data[-1, 1, 1] = abs(fx)
         data[-1, 1, 2] = abs(fy)
         # Moving average on force data (column 1): simple loop, no singularities at ends

src/orisys/sensor.py

@@ -12,8 +12,8 @@ import cv2
 import time
 from enum import Enum, auto
 from typing import Tuple
-from threading import Thread, Event
+# from threading import Thread, Event
-from queue import Queue, Empty
+# from queue import Queue, Empty
 
 from .Stitcher import Stitcher
 from .tracker import Tracker

src/orisys/tracker.py

@@ -25,11 +25,15 @@ class Tracker:
         dis_preset = config_tracker["dis_preset"]
         loop_closure_interval = config_tracker["loop_closure_interval"]
         self.__algorithm = config_tracker["algorithm"]
+        
+        # Conditionally import torch only if RAFT algorithm is used
+        self.torch = None
         if self.__algorithm == "raft":
             import torch
+            self.torch = torch
 
 
-        if self.__algorithm == "raft" and (not self.cuda or not torch.cuda.is_available()):
+        if self.__algorithm == "raft" and (not self.cuda or not self.torch.cuda.is_available()):
             self.__algorithm = "dis"
             if self.verbose:
                 print("No CUDA available, falling back to OPENCV DIS optical flow")
@@ -217,7 +221,7 @@ class Tracker:
             frame = np.ascontiguousarray(frame)
         
         # Upload to device as uint8 first to save bandwidth (4x less data than float32)
-        tensor = torch.from_numpy(frame).permute(2, 0, 1)
+        tensor = self.torch.from_numpy(frame).permute(2, 0, 1)
         
         # Pin memory for faster CPU-GPU transfer (similar to DataLoader pin_memory=True)
         if self.__raft_device == 'cuda':
@@ -258,9 +262,9 @@ class Tracker:
         if self.verbose:
             t0 = time.perf_counter()
 
-        with torch.inference_mode():
+        with self.torch.inference_mode():
             if self.__raft_device == 'cuda':
-                with torch.amp.autocast('cuda'):
+                with self.torch.amp.autocast('cuda'):
                     flow = self.__raft_model(prev_tensor, curr_tensor)
             else:
                 flow = self.__raft_model(prev_tensor, curr_tensor)