Run object detection model on the Raspberry Pi

Overview

Intro

Using TensorFlow Lite with Python is great for embedded devices based on Linux, such as Raspberry Pi.

This is the guide for installing TensorFlow Lite on the Raspberry Pi and running pre-trained object detection models on it.

Step 1. Setting up Rasperry Pi

Upgrade Raspbian Stretch to Buster

(If you on Buster, skip this step and simply run sudo apt-get update and sudo apt-get dist-upgrade)

$ sudo apt-get update && sudo apt-get upgrade -y

Verify nothing is wrong. Verify no errors are reported after each command. Fix as required (you’re on your own here!).

$ dpkg -C
$ apt-mark showhold

Prepare apt-get Sources

Update the sources to apt-get. This replaces “stretch” with “buster” in the repository locations giving apt-get access to the new version’s binaries.

$ sudo sed -i 's/stretch/buster/g' /etc/apt/sources.list    
$ sudo sed -i 's/stretch/buster/g' /etc/apt/sources.list.d/raspi.list

Verify this caught them all by running the following, expecting no output. If the command returns anything having previously run the sed commands above, it means more files may need tweaking. Run the sed command for each. The aim is to replace all instances of “stretch”.

$ grep -lnr stretch /etc/apt

Speed up subsequent steps by removing the list change package.

$ sudo apt-get remove apt-listchanges

Do the Upgrade

To update existing packages without updating kernel modules or removing packages, run the following.

$ sudo apt-get update && sudo apt-get upgrade -y

Alternatively, to include kernel modules and removing packages if required, run the following

$ sudo apt-get update && sudo apt-get full-upgrade -y

Cleanup old outdated packages.

$ sudo apt-get autoremove -y && sudo apt-get autoclean

Verify with

 cat /etc/os-release.

Update Firmware

$ sudo rpi-update

and

sudo apt-get install -y python3-pip

and

pip3 install --upgrade setuptools

2. Making sure camera interface is enabled in the Raspberry Pi Configuration menu

Click the Pi icon in the top left corner of the screen, select Preferences -> Raspberry Pi Configuration, and go to the Interfaces tab and verify Camera is set to Enabled. If it isn't, enable it now, and reboot the Raspberry Pi.

Converting Tensorflow to Tensorflow Lite

Using TensorFlow Lite converter. It takes TensorFlow model and generates a TensorFlow Lite model (an optimized FlatBuffer format identified by the .tflite file extension).

Step 2. Install TF Lite dependecies and set up virtual environment

clone this repo

git clone https://github.com/yanovsk/Raspberry-Pi-TF-Lite-Object-Detection

rename the folder to "tfliteod"

mv Raspberry-Pi-TF-Lite-Object-Detection tfliteod
cd tfliteod

run shell script to install get_pi_requirements

bash get_pi_req.sh

Note: shell script will auto install the lastest version of Tensorflow. To install specific version of TF, run pip3 install tensorflow==x.xx (where x.xx stands for the version you want to install)

Set up virtual environment

Install vitrtualenv

pip3 install virtualenv 

Then, create the "tfliteod-env" virtual environment by issuing:

python3 -m venv tfliteod-env

This will create a folder called tfliteod-env inside the tflite1 directory. The tfliteod-env folder will hold all the package libraries for this environment. Next, activate the environment by issuing:

source tfliteod-env/bin/activate

Step 3. Set up TensorFlow Lite detection model

Once, tensorflow is install we can proceed to seting up the object detection model.

We can use either pre-trained model or train it on our end. For the simplicity sake let's use pre-trained sample model by google

Download the sample model (also could be done thru direct link here)

wget https://storage.googleapis.com/download.tensorflow.org/models/tflite/coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip

upzip it

unzip coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip -d Sample_model

Step 4. Run the model

Note: the model should work on either Picamera module or any other webcam plugged in to the Raspberry Pi as a usb device.

From home/pi/tfliteod run the following command:

python3 TFL_object_detection.py --modeldir=Sample_model

After initializing webcam window should pop-up on your Raspebbery Pi and object detection should work.

Note: this model can recongnize only 80 common objects (check labels.txt for more info on metadata)

However, you can custom train the model using this guide.

Happy hacking!

Owner
Dimitri Yanovsky
Dimitri Yanovsky
[EMNLP 2021] MuVER: Improving First-Stage Entity Retrieval with Multi-View Entity Representations

MuVER This repo contains the code and pre-trained model for our EMNLP 2021 paper: MuVER: Improving First-Stage Entity Retrieval with Multi-View Entity

24 May 30, 2022
The official repository for "Intermediate Layers Matter in Momentum Contrastive Self Supervised Learning" paper.

Intermdiate layer matters - SSL The official repository for "Intermediate Layers Matter in Momentum Contrastive Self Supervised Learning" paper. Downl

Aakash Kaku 35 Sep 19, 2022
MMdet2-based reposity about lightweight detection model: Nanodet, PicoDet.

Lightweight-Detection-and-KD MMdet2-based reposity about lightweight detection model: Nanodet, PicoDet. This repo also includes detection knowledge di

Egqawkq 12 Jan 05, 2023
MEDS: Enhancing Memory Error Detection for Large-Scale Applications

MEDS: Enhancing Memory Error Detection for Large-Scale Applications Prerequisites cmake and clang Build MEDS supporting compiler $ make Build Using Do

Secomp Lab at Purdue University 34 Dec 14, 2022
Simulations for Turring patterns on an apically expanding domain. T

Turing patterns on expanding domain Simulations for Turring patterns on an apically expanding domain. The details about the models and numerical imple

Yue Liu 0 Aug 03, 2021
[Link]deep_portfolo - Use Reforcemet earg ad Supervsed learg to Optmze portfolo allocato []

rl_portfolio This Repository uses Reinforcement Learning and Supervised learning to Optimize portfolio allocation. The goal is to make profitable agen

Deepender Singla 165 Dec 02, 2022
Data Augmentation Using Keras and Python

Data-Augmentation-Using-Keras-and-Python Data augmentation is the process of increasing the number of training dataset. Keras library offers a simple

Happy N. Monday 3 Feb 15, 2022
This repository contains pre-trained models and some evaluation code for our paper Towards Unsupervised Dense Information Retrieval with Contrastive Learning

Contriever: Towards Unsupervised Dense Information Retrieval with Contrastive Learning This repository contains pre-trained models and some evaluation

Meta Research 207 Jan 08, 2023
Code and models for "Pano3D: A Holistic Benchmark and a Solid Baseline for 360 Depth Estimation", OmniCV Workshop @ CVPR21.

Pano3D A Holistic Benchmark and a Solid Baseline for 360o Depth Estimation Pano3D is a new benchmark for depth estimation from spherical panoramas. We

Visual Computing Lab, Information Technologies Institute, Centre for Reseach and Technology Hellas 50 Dec 29, 2022
Seeing Dynamic Scene in the Dark: High-Quality Video Dataset with Mechatronic Alignment (ICCV2021)

Seeing Dynamic Scene in the Dark: High-Quality Video Dataset with Mechatronic Alignment This is a pytorch project for the paper Seeing Dynamic Scene i

DV Lab 21 Nov 28, 2022
Related resources for our EMNLP 2021 paper

Plan-then-Generate: Controlled Data-to-Text Generation via Planning Authors: Yixuan Su, David Vandyke, Sihui Wang, Yimai Fang, and Nigel Collier Code

Yixuan Su 61 Jan 03, 2023
MNIST, but with Bezier curves instead of pixels

bezier-mnist This is a work-in-progress vector version of the MNIST dataset. Samples Here are some samples from the training set. Note that, while the

Alex Nichol 15 Jan 16, 2022
Official Implementation of LARGE: Latent-Based Regression through GAN Semantics

LARGE: Latent-Based Regression through GAN Semantics [Project Website] [Google Colab] [Paper] LARGE: Latent-Based Regression through GAN Semantics Yot

83 Dec 06, 2022
dyld_shared_cache processing / Single-Image loading for BinaryNinja

Dyld Shared Cache Parser Author: cynder (kat) Dyld Shared Cache Support for BinaryNinja Without any of the fuss of requiring manually loading several

cynder 76 Dec 28, 2022
Using PyTorch Perform intent classification using three different models to see which one is better for this task

Using PyTorch Perform intent classification using three different models to see which one is better for this task

Yoel Graumann 1 Feb 14, 2022
Data from "HateCheck: Functional Tests for Hate Speech Detection Models" (Röttger et al., ACL 2021)

In this repo, you can find the data from our ACL 2021 paper "HateCheck: Functional Tests for Hate Speech Detection Models". "test_suite_cases.csv" con

Paul Röttger 43 Nov 11, 2022
A lane detection integrated Real-time Instance Segmentation based on YOLACT (You Only Look At CoefficienTs)

Real-time Instance Segmentation and Lane Detection This is a lane detection integrated Real-time Instance Segmentation based on YOLACT (You Only Look

Jin 4 Dec 30, 2022
Learning Calibrated-Guidance for Object Detection in Aerial Images

Learning Calibrated-Guidance for Object Detection in Aerial Images arxiv We propose a simple yet effective Calibrated-Guidance (CG) scheme to enhance

51 Sep 22, 2022
GNN-based Recommendation Benchmark

GRecX A Fair Benchmark for GNN-based Recommendation Homepage and Documentation Homepage: Documentation: Paper: GRecX: An Efficient and Unified Benchma

73 Oct 17, 2022
Learning from Guided Play: A Scheduled Hierarchical Approach for Improving Exploration in Adversarial Imitation Learning Source Code

Learning from Guided Play: A Scheduled Hierarchical Approach for Improving Exploration in Adversarial Imitation Learning Source Code

STARS Laboratory 8 Sep 14, 2022