diff --git a/README.md b/README.md
index b883612..b9a73c9 100644
--- a/README.md
+++ b/README.md
@@ -41,7 +41,7 @@ Execute this command using `sudo` if you are not using a virtual environment.
 
 ## How to Use
 
-You need to use a TensorFlow neural network on which running the inferences. We provide in this repository a pre-trained model taken from [the TensorFlow Model Zoo](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md). There you can find more models ready to work. You can choose a different one, but make sure that it outputs __boxes__ (masks are not supported yet), and specify its dataset and the path to the `.pb` file containing the graph in `objectdetector.yml`.
+You need to use a TensorFlow neural network on which running the inferences. We provide in this repository a pre-trained model taken from [the TensorFlow Model Zoo](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md). There you can find more models ready to work. You can choose a different one, but make sure that it outputs __boxes__ (masks are not supported yet), and specify its dataset and the path to the `.pb` file containing the graph in `objectdetector.yml`.
 
 
 #### Detections from a video stream (ICE/ROS):
diff --git a/requirements.txt b/requirements.txt
index 175d80f..39693c0 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,5 +1,5 @@
-numpy==1.14 # to avoid deprecation warnings
-tensorflow==1.14.0
-opencv-python
-Pillow
-cprint
\ No newline at end of file
+numpy # to avoid deprecation warnings
+tensorflow
+opencv-python # if it doesn't work, install from source - "sudo apt install python3-opencv"
+Pillow==2.2.2
+cprint