diff --git a/src/detector.c b/src/detector.c
index 5f2cec83..f401d20a 100644
--- a/src/detector.c
+++ b/src/detector.c
@@ -70,6 +70,9 @@ void train_detector(char *datacfg, char *cfgfile, char *weightfile, int *gpus, i
     //int N = plist->size;
     char **paths = (char **)list_to_array(plist);
 
+	int init_w = net.w;
+	int init_h = net.h;
+
     load_args args = {0};
     args.w = net.w;
     args.h = net.h;
@@ -95,8 +98,9 @@ void train_detector(char *datacfg, char *cfgfile, char *weightfile, int *gpus, i
     while(get_current_batch(net) < net.max_batches){
 		if(l.random && count++%10 == 0){
             printf("Resizing\n");
-            int dim = (rand() % 10 + 10) * 32;
-            if (get_current_batch(net)+100 > net.max_batches) dim = 544;
+			int dim = (rand() % 12 + (init_w/32 - 5)) * 32;	// +-160
+            //int dim = (rand() % 10 + 10) * 32;
+            //if (get_current_batch(net)+100 > net.max_batches) dim = 544;
             //int dim = (rand() % 4 + 16) * 32;
             printf("%d\n", dim);
             args.w = dim;
@@ -152,7 +156,8 @@ void train_detector(char *datacfg, char *cfgfile, char *weightfile, int *gpus, i
 
         i = get_current_batch(net);
         printf("%d: %f, %f avg, %f rate, %lf seconds, %d images\n", get_current_batch(net), loss, avg_loss, get_current_rate(net), sec(clock()-time), i*imgs);
-		if (i % 1000 == 0 || (i < 1000 && i % 100 == 0)) {
+		//if (i % 1000 == 0 || (i < 1000 && i % 100 == 0)) {
+		if (i % 100 == 0) {
 #ifdef GPU
 			if (ngpus != 1) sync_nets(nets, ngpus, 0);
 #endif
diff --git a/src/network_kernels.cu b/src/network_kernels.cu
index 9ecabdf3..64f4f9b8 100644
--- a/src/network_kernels.cu
+++ b/src/network_kernels.cu
@@ -114,6 +114,7 @@ void forward_backward_network_gpu(network net, float *x, float *y)
     state.truth = *net.truth_gpu;
     state.train = 1;
     forward_network_gpu(net, state);
+	cudaStreamSynchronize(get_cuda_stream());
     backward_network_gpu(net, state);
 }
 
diff --git a/src/region_layer.c b/src/region_layer.c
index 0638301f..7772bc30 100644
--- a/src/region_layer.c
+++ b/src/region_layer.c
@@ -419,6 +419,7 @@ void forward_region_layer_gpu(const region_layer l, network_state state)
     free(cpu_state.input);
     if(!state.train) return;
     cuda_push_array(l.delta_gpu, l.delta, l.batch*l.outputs);
+	cudaStreamSynchronize(get_cuda_stream());
     if(cpu_state.truth) free(cpu_state.truth);
 }
 
diff --git a/src/yolo_console_dll.cpp b/src/yolo_console_dll.cpp
index 5acca7cf..a9ce6b9e 100644
--- a/src/yolo_console_dll.cpp
+++ b/src/yolo_console_dll.cpp
@@ -194,7 +194,7 @@ int main(int argc, char *argv[])
 							auto current_image = det_image;
 							consumed = true;
 							while (current_image.use_count() > 0) {
-								auto result = detector.detect_resized(*current_image, frame_size, 0.24, false);	// true
+								auto result = detector.detect_resized(*current_image, frame_size, 0.20, false);	// true
 								++fps_det_counter;
 								std::unique_lock<std::mutex> lock(mtx);
 								thread_result_vec = result;
@@ -236,11 +236,13 @@ int main(int argc, char *argv[])
 						}
 					}
 
+#ifndef TRACK_OPTFLOW
 					// wait detection result for video-file only (not for net-cam)
-					//if (protocol != "rtsp://" && protocol != "http://" && protocol != "https:/") {
-					//	std::unique_lock<std::mutex> lock(mtx);
-					//	while (!consumed) cv_detected.wait(lock);
-					//}
+					if (protocol != "rtsp://" && protocol != "http://" && protocol != "https:/") {
+						std::unique_lock<std::mutex> lock(mtx);
+						while (!consumed) cv_detected.wait(lock);
+					}
+#endif
 				}
 				if (t_cap.joinable()) t_cap.join();
 				if (t_detect.joinable()) t_detect.join();