From fb1208c16ca9af7f01ce8abfbf9a502b06a36865 Mon Sep 17 00:00:00 2001
From: Raul James <raul@nobi.io>
Date: Thu, 29 Feb 2024 13:59:42 +0100
Subject: [PATCH] step through video frame_interval at a time

---
 examples/image/classify-video.py | 58 ++++++++++++++++++++++++--------
 1 file changed, 44 insertions(+), 14 deletions(-)

diff --git a/examples/image/classify-video.py b/examples/image/classify-video.py
index cdea9b5..649a944 100644
--- a/examples/image/classify-video.py
+++ b/examples/image/classify-video.py
@@ -12,12 +12,13 @@
 runner = None
 # if you don't want to see a video preview, set this to False
 show_camera = True
+
 if (sys.platform == 'linux' and not os.environ.get('DISPLAY')):
     show_camera = False
 
 
 def help():
-    print('python classify-video.py <path_to_model.eim> <path_to_video.mp4>')
+    print('python classify-video.py <path_to_model.eim> <path_to_video.mp4> [frame_interval]')
 
 def main(argv):
     try:
@@ -31,53 +32,80 @@ def main(argv):
             help()
             sys.exit()
 
-    if len(args) != 2:
+    if len(args) < 2 or len(args) > 3:
         help()
         sys.exit(2)
 
     model = args[0]
+    video_file = args[1]
+
 
     dir_path = os.path.dirname(os.path.realpath(__file__))
     modelfile = os.path.join(dir_path, model)
-
     print('MODEL: ' + modelfile)
 
+    # jump 1 frame at a time by default
+    frame_interval = 1
+
+    if len(args) == 3:
+        try:
+            frame_interval = int(args[2])
+        except ValueError:
+            frame_interval = None
+    print('Stepping through %d frames at a time' % frame_interval)
+
+
     with ImageImpulseRunner(modelfile) as runner:
         try:
             model_info = runner.init()
             print('Loaded runner for "' + model_info['project']['owner'] + ' / ' + model_info['project']['name'] + '"')
             labels = model_info['model_parameters']['labels']
 
-            vidcap = cv2.VideoCapture(args[1])
-            sec = 0
+            vidcap = cv2.VideoCapture(video_file)
+            #fps = vidcap.get(cv2.CAP_PROP_FPS)
+            #total_frames = vidcap.get(cv2.CAP_PROP_FRAME_COUNT)
+            #total_duration = total_frames / fps
+            #print("File information: FPS: %.2f Frames %d, duration: %.2f" % (fps, total_frames, total_duration))
+
+            frame_nr = 0
             start_time = time.time()
 
-            def getFrame(sec):
-                vidcap.set(cv2.CAP_PROP_POS_MSEC,sec*1000)
+            def getFrame(id):
+                #debug_time = time.time()
+                #print("frame id: %d" % id)
+                vidcap.set(cv2.CAP_PROP_POS_FRAMES,id)
+
                 hasFrames,image = vidcap.read()
                 if hasFrames:
+                    #print("get frame took: %.2f sec" % (time.time() - debug_time))
                     return image
                 else:
-                    print('Failed to load frame', args[1])
+                    print('Failed to load frame', video_file)
                     exit(1)
 
 
-            img = getFrame(sec)
+
+            img = getFrame(frame_nr)
 
             while img.size != 0:
 
                 # imread returns images in BGR format, so we need to convert to RGB
+                #debug_time  = time.time()
                 img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+                #print("color convert took: %.2f sec" % (time.time() - debug_time))
 
                 # get_features_from_image also takes a crop direction arguments in case you don't have square images
+                #debug_time  = time.time()
                 features, cropped = runner.get_features_from_image(img)
+                #print("get features took: %.2f sec" % (time.time() - debug_time))
 
                 # the image will be resized and cropped, save a copy of the picture here
                 # so you can see what's being passed into the classifier
-                cv2.imwrite('debug.jpg', cv2.cvtColor(cropped, cv2.COLOR_RGB2BGR))
+                #cv2.imwrite('debug.jpg', cv2.cvtColor(cropped, cv2.COLOR_RGB2BGR))
 
                 res = runner.classify(features)
 
+                #debug_time  = time.time()
                 if "classification" in res["result"].keys():
                     print('Result (%d ms.) ' % (res['timing']['dsp'] + res['timing']['classification']), end='')
                     for label in labels:
@@ -90,16 +118,18 @@ def getFrame(sec):
                     for bb in res["result"]["bounding_boxes"]:
                         print('\t%s (%.2f): x=%d y=%d w=%d h=%d' % (bb['label'], bb['value'], bb['x'], bb['y'], bb['width'], bb['height']))
                         img = cv2.rectangle(cropped, (bb['x'], bb['y']), (bb['x'] + bb['width'], bb['y'] + bb['height']), (255, 0, 0), 1)
+                #print("process results took: %.2f sec" % (time.time() - debug_time))
 
                 if (show_camera):
+                    #debug_time = time.time()
                     cv2.imshow('edgeimpulse', cv2.cvtColor(cropped, cv2.COLOR_RGB2BGR))
                     if cv2.waitKey(1) == ord('q'):
                         break
+                    #print("show results took: %.2f sec" % (time.time() - debug_time))
+
 
-                sec = time.time() - start_time
-                sec = round(sec, 2)
-                print("Getting frame at: %.2f sec" % sec)
-                img = getFrame(sec)
+                frame_nr += frame_interval
+                img = getFrame(frame_nr)
         finally:
             if (runner):
                 runner.stop()