changes made for compatibility with python3

annotation-tool.py

@@ -52,7 +52,7 @@ def distanceTo(self,pt):
 
 	def shiftPts(self):
 		if self.pts.shape[1] > 1:
-			idx = range(1,self.pts.shape[1]) + [0]
+			idx = list(range(1,self.pts.shape[1])) + [0]
 			self.pts = self.pts[...,idx]
 
 	def getSquare(self):
@@ -227,7 +227,7 @@ def displayAllShapes(disp,shapes,selected,typing_mode):
 if __name__ == '__main__':
 
 	if len(sys.argv) < 4:
-		print __doc__
+		print(__doc__)
 		sys.exit()
 
 	maxW = int(sys.argv[1])
@@ -343,19 +343,19 @@ def displayAllShapes(disp,shapes,selected,typing_mode):
 					typing_mode = True
 
 				if key == key_append_vertex:
-					print 'Append vertex'
+					print('Append vertex')
 					shapes[selected].appendSide(disp.getMouseCenterRelative())
 
 				if key == key_remove_last_vertex:
-					print 'Remove last vertex'
+					print('Remove last vertex')
 					shapes[selected].removeLast()
 
 				if key == key_change_closest_vertex:
-					print 'Change closest vertex'
+					print('Change closest vertex')
 					shapes[selected].changeClosest(disp.getMouseCenterRelative())
 
 				if key in key_delete_selected_shape:
-					print 'Delete closest vertex'
+					print('Delete closest vertex')
 					del shapes[selected]
 					pt = disp.getMouseCenterRelative()
 					selected = selectClosest(shapes,pt)
@@ -364,12 +364,12 @@ def displayAllShapes(disp,shapes,selected,typing_mode):
 					shapes[selected].shiftPts()
 
 			if key == key_create_new_shape:
-				print 'Create new shape'
+				print('Create new shape')
 				shapes.append(ShapeDisplay())
 				selected = len(shapes)-1
 
 			if key == key_select_closest_shape:
-				print 'Select closest'
+				print('Select closest')
 				pt = disp.getMouseCenterRelative()
 				selected = selectClosest(shapes,pt)
 

create-model.py

@@ -83,7 +83,7 @@ def create_model_mobnet():
 	backbone_layers = {'backbone_' + layer.name: layer for layer in backbone.layers}
 	for layer in model.layers:
 		if layer.name in backbone_layers:
-			print 'setting ' + layer.name
+			print(('setting ' + layer.name))
 			layer.set_weights(backbone_layers[layer.name].get_weights())
 
 	return model
@@ -98,10 +98,10 @@ def create_model_mobnet():
 
 	modelf = getattr(sys.modules[__name__],'create_model_' + sys.argv[1])
 
-	print 'Creating model %s' % sys.argv[1]
+	print(('Creating model %s' % sys.argv[1]))
 	model = modelf()
-	print 'Finished'
+	print('Finished')
 
-	print 'Saving at %s' % sys.argv[2]
+	print(('Saving at %s' % sys.argv[2]))
 	save_model(model,sys.argv[2])
 

darknet/examples/detector-scipy-opencv.py

@@ -39,18 +39,18 @@ def detect2(net, meta, image, thresh=.5, hier_thresh=.5, nms=.45):
 net = dn.load_net("cfg/tiny-yolo.cfg", "tiny-yolo.weights", 0)
 meta = dn.load_meta("cfg/coco.data")
 r = dn.detect(net, meta, "data/dog.jpg")
-print r
+print(r)
 
 # scipy
 arr= imread('data/dog.jpg')
 im = array_to_image(arr)
 r = detect2(net, meta, im)
-print r
+print(r)
 
 # OpenCV
 arr = cv2.imread('data/dog.jpg')
 im = array_to_image(arr)
 dn.rgbgr_image(im)
 r = detect2(net, meta, im)
-print r
+print(r)
 

darknet/examples/detector.py

@@ -13,15 +13,15 @@
 net = dn.load_net("cfg/yolo-thor.cfg", "/home/pjreddie/backup/yolo-thor_final.weights", 0)
 meta = dn.load_meta("cfg/thor.data")
 r = dn.detect(net, meta, "data/bedroom.jpg")
-print r
+print(r)
 
 # And then down here you could detect a lot more images like:
 r = dn.detect(net, meta, "data/eagle.jpg")
-print r
+print(r)
 r = dn.detect(net, meta, "data/giraffe.jpg")
-print r
+print(r)
 r = dn.detect(net, meta, "data/horses.jpg")
-print r
+print(r)
 r = dn.detect(net, meta, "data/person.jpg")
-print r
+print(r)
 

darknet/python/darknet.py

@@ -123,6 +123,7 @@ def classify(net, meta, im):
     return res
 
 def detect(net, meta, image, thresh=.5, hier_thresh=.5, nms=.45):
+    image = bytes(image, encoding='utf-8')
     im = load_image(image, 0, 0)
     num = c_int(0)
     pnum = pointer(num)
@@ -152,6 +153,6 @@ def detect(net, meta, image, thresh=.5, hier_thresh=.5, nms=.45):
     net = load_net("cfg/tiny-yolo.cfg", "tiny-yolo.weights", 0)
     meta = load_meta("cfg/coco.data")
     r = detect(net, meta, "data/dog.jpg")
-    print r
+    print(r)
 
 

darknet/python/proverbot.py

@@ -1,4 +1,4 @@
-from darknet import *
+from .darknet import *
 
 def predict_tactic(net, s):
     prob = 0
@@ -34,4 +34,4 @@ def predict_tactics(net, s, n):
 
 net = load_net("cfg/coq.test.cfg", "/home/pjreddie/backup/coq.backup", 0)
 t = predict_tactics(net, "+++++\n", 10)
-print t
+print(t)

get-networks.sh

@@ -6,15 +6,15 @@ mkdir data/lp-detector -p
 mkdir data/ocr -p
 mkdir data/vehicle-detector -p
 
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/lp-detector/wpod-net_update1.h5   -P data/lp-detector/
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/lp-detector/wpod-net_update1.json -P data/lp-detector/
+wget -c -N http://sergiomsilva.com/data/eccv2018/lp-detector/wpod-net_update1.h5   -P data/lp-detector/
+wget -c -N http://sergiomsilva.com/data/eccv2018/lp-detector/wpod-net_update1.json -P data/lp-detector/
 
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/ocr/ocr-net.cfg     -P data/ocr/
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/ocr/ocr-net.names   -P data/ocr/
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/ocr/ocr-net.weights -P data/ocr/
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/ocr/ocr-net.data    -P data/ocr/
+wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.cfg     -P data/ocr/
+wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.names   -P data/ocr/
+wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.weights -P data/ocr/
+wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.data    -P data/ocr/
 
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/vehicle-detector/yolo-voc.cfg     -P data/vehicle-detector/
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/vehicle-detector/voc.data         -P data/vehicle-detector/
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/vehicle-detector/yolo-voc.weights -P data/vehicle-detector/
-wget -c -N www.inf.ufrgs.br/~smsilva/alpr-unconstrained/data/vehicle-detector/voc.names        -P data/vehicle-detector/
+wget -c -N http://sergiomsilva.com/data/eccv2018/vehicle-detector/yolo-voc.cfg     -P data/vehicle-detector/
+wget -c -N http://sergiomsilva.com/data/eccv2018/vehicle-detector/voc.data         -P data/vehicle-detector/
+wget -c -N http://sergiomsilva.com/data/eccv2018/vehicle-detector/yolo-voc.weights -P data/vehicle-detector/
+wget -c -N http://sergiomsilva.com/data/eccv2018/vehicle-detector/voc.names        -P data/vehicle-detector/

license-plate-detection.py

@@ -29,19 +29,19 @@ def adjust_pts(pts,lroi):
 
 		imgs_paths = glob('%s/*car.png' % input_dir)
 
-		print 'Searching for license plates using WPOD-NET'
+		print('Searching for license plates using WPOD-NET')
 
 		for i,img_path in enumerate(imgs_paths):
 
-			print '\t Processing %s' % img_path
+			print(('\t Processing %s' % img_path))
 
 			bname = splitext(basename(img_path))[0]
 			Ivehicle = cv2.imread(img_path)
 
 			ratio = float(max(Ivehicle.shape[:2]))/min(Ivehicle.shape[:2])
 			side  = int(ratio*288.)
 			bound_dim = min(side + (side%(2**4)),608)
-			print "\t\tBound dim: %d, ratio: %f" % (bound_dim,ratio)
+			print(("\t\tBound dim: %d, ratio: %f" % (bound_dim,ratio)))
 
 			Llp,LlpImgs,_ = detect_lp(wpod_net,im2single(Ivehicle),bound_dim,2**4,(240,80),lp_threshold)
 

license-plate-ocr.py

@@ -21,20 +21,20 @@
 
 		ocr_threshold = .4
 
-		ocr_weights = 'data/ocr/ocr-net.weights'
-		ocr_netcfg  = 'data/ocr/ocr-net.cfg'
-		ocr_dataset = 'data/ocr/ocr-net.data'
+		ocr_weights = b'data/ocr/ocr-net.weights'
+		ocr_netcfg  = b'data/ocr/ocr-net.cfg'
+		ocr_dataset = b'data/ocr/ocr-net.data'
 
 		ocr_net  = dn.load_net(ocr_netcfg, ocr_weights, 0)
 		ocr_meta = dn.load_meta(ocr_dataset)
 
 		imgs_paths = sorted(glob('%s/*lp.png' % output_dir))
 
-		print 'Performing OCR...'
+		print('Performing OCR...')
 
 		for i,img_path in enumerate(imgs_paths):
 
-			print '\tScanning %s' % img_path
+			print(('\tScanning %s' % img_path))
 
 			bname = basename(splitext(img_path)[0])
 
@@ -51,11 +51,11 @@
 				with open('%s/%s_str.txt' % (output_dir,bname),'w') as f:
 					f.write(lp_str + '\n')
 
-				print '\t\tLP: %s' % lp_str
+				print(('\t\tLP: %s' % lp_str))
 
 			else:
 
-				print 'No characters found'
+				print('No characters found')
 
 	except:
 		traceback.print_exc()

src/keras_utils.py

@@ -24,7 +24,7 @@ def save_model(model,path,verbose=0):
 	with open('%s.json' % path,'w') as json_file:
 		json_file.write(model_json)
 	model.save_weights('%s.h5' % path)
-	if verbose: print 'Saved to %s' % path
+	if verbose: print(('Saved to %s' % path))
 
 def load_model(path,custom_objects={},verbose=0):
 	from keras.models import model_from_json
@@ -34,7 +34,7 @@ def load_model(path,custom_objects={},verbose=0):
 		model_json = json_file.read()
 	model = model_from_json(model_json, custom_objects=custom_objects)
 	model.load_weights('%s.h5' % path)
-	if verbose: print 'Loaded from %s' % path
+	if verbose: print(('Loaded from %s' % path))
 	return model
 
 

train-detector.py

@@ -76,7 +76,7 @@ def process_data_item(data_item,dim,model_stride):
 	opt = getattr(keras.optimizers,args.optimizer)(lr=args.learning_rate)
 	model.compile(loss=loss, optimizer=opt)
 
-	print 'Checking input directory...'
+	print('Checking input directory...')
 	Files = image_files_from_folder(train_dir)
 
 	Data = []
@@ -87,7 +87,7 @@ def process_data_item(data_item,dim,model_stride):
 			I = cv2.imread(file)
 			Data.append([I,L[0]])
 
-	print '%d images with labels found' % len(Data)
+	print(('%d images with labels found' % len(Data)))
 
 	dg = DataGenerator(	data=Data, \
 						process_data_item_func=lambda x: process_data_item(x,dim,model_stride),\
@@ -106,20 +106,20 @@ def process_data_item(data_item,dim,model_stride):
 
 	for it in range(iterations):
 
-		print 'Iter. %d (of %d)' % (it+1,iterations)
+		print(('Iter. %d (of %d)' % (it+1,iterations)))
 
 		Xtrain,Ytrain = dg.get_batch(batch_size)
 		train_loss = model.train_on_batch(Xtrain,Ytrain)
 
-		print '\tLoss: %f' % train_loss
+		print(('\tLoss: %f' % train_loss))
 
 		# Save model every 1000 iterations
 		if (it+1) % 1000 == 0:
-			print 'Saving model (%s)' % model_path_backup
+			print(('Saving model (%s)' % model_path_backup))
 			save_model(model,model_path_backup)
 
-	print 'Stopping data generator'
+	print('Stopping data generator')
 	dg.stop()
 
-	print 'Saving model (%s)' % model_path_final
+	print(('Saving model (%s)' % model_path_final))
 	save_model(model,model_path_final)

vehicle-detection.py

@@ -21,10 +21,16 @@
 
 		vehicle_threshold = .5
 
-		vehicle_weights = 'data/vehicle-detector/yolo-voc.weights'
-		vehicle_netcfg  = 'data/vehicle-detector/yolo-voc.cfg'
-		vehicle_dataset = 'data/vehicle-detector/voc.data'
+		base=b'/media/raghav/e34065bb-bd49-4111-ba3a-96160e27ffd0/raghu/cctv/darknet/'
+		vehicle_weights = base + b'yolov4.weights'
+		vehicle_netcfg  = base + b'cfg/yolov4.cfg'
+		vehicle_dataset = b'data/vehicle-detector/coco.data'
 
+		vehicle_weights = b'data/vehicle-detector/yolo-voc.weights'
+		vehicle_netcfg  = b'data/vehicle-detector/yolo-voc.cfg'
+		vehicle_dataset = b'data/vehicle-detector/voc.data'
+
+		# import pdb; pdb.set_trace()
 		vehicle_net  = dn.load_net(vehicle_netcfg, vehicle_weights, 0)
 		vehicle_meta = dn.load_meta(vehicle_dataset)
 
@@ -34,19 +40,19 @@
 		if not isdir(output_dir):
 			makedirs(output_dir)
 
-		print 'Searching for vehicles using YOLO...'
+		print('Searching for vehicles using YOLO...')
 
 		for i,img_path in enumerate(imgs_paths):
 
-			print '\tScanning %s' % img_path
+			print(('\tScanning %s' % img_path))
 
 			bname = basename(splitext(img_path)[0])
 
 			R,_ = detect(vehicle_net, vehicle_meta, img_path ,thresh=vehicle_threshold)
 
-			R = [r for r in R if r[0] in ['car','bus']]
+			R = [r for r in R if r[0] in [b'car',b'bus']]
 
-			print '\t\t%d cars found' % len(R)
+			print(('\t\t%d cars found' % len(R)))
 
 			if len(R):