using Neural Networks (SSD) on Tensorflow. This repo documents steps and scripts used to train a hand detector using Tensorflow (Object Detection API). As with any DNN based task, the most expensive (and riskiest) part of the process has to do with finding or creating the right (annotated) dataset. I was interested mainly in detecting hands on a table (egocentric view point). I experimented first with the [Oxford Hands Dataset](http://www.robots.ox.ac.uk/~vgg/data/hands/) (the results were not good). I then tried the [Egohands Dataset](http://vision.soic.indiana.edu/projects/egohands/) which was a much better fit to my requirements. The goal of this repo/post is to demonstrate how neural networks can be applied to the (hard) problem of tracking hands (egocentric and other views). Better still, provide code that can be adapted to other uses cases. If you use this tutorial or models in your research or project, please cite [this](#citing-this-tutorial). Here is the detector in action. <img src="images/hand1.gif" width="33.3%"><img src="images/hand2.gif" width="33.3%"><img src="images/hand3.gif" width="33.3%"> Realtime detection on video stream from a webcam . <img src="images/chess1.gif" width="33.3%"><img src="images/chess2.gif" width="33.3%"><img src="images/chess3.gif" width="33.3%"> Detection on a Youtube video. Both examples above were run on a macbook pro **CPU** (i7, 2.5GHz, 16GB). Some fps numbers are: | FPS | Image Size | Device| Comments| | ------------- | ------------- | ------------- | ------------- | | 21 | 320 * 240 | Macbook pro (i7, 2.5GHz, 16GB) | Run without visualizing results| | 16 | 320 * 240 | Macbook pro (i7, 2.5GHz, 16GB) | Run while visualizing results (image above) | | 11 | 640 * 480 | Macbook pro (i7, 2.5GHz, 16GB) | Run while visualizing results (image above) | > Note: The code in this repo is written and tested with Tensorflow `1.4.0-rc0`. Using a different version may result in [some errors](https://github.com/tensorflow/models/issues/1581). You may need to [generate your own frozen model](https://pythonprogramming.net/testing-custom-object-detector-tensorflow-object-detection-api-tutorial/?completed=/training-custom-objects-tensorflow-object-detection-api-tutorial/) graph using the [model checkpoints](model-checkpoint) in the repo to fit your TF version. **Content of this document** - Motivation - Why Track/Detect hands with Neural Networks - Data preparation and network training in Tensorflow (Dataset, Import, Training) - Training the hand detection Model - Using the Detector to Detect/Track hands - Thoughts on Optimizations. > P.S if you are using or have used the models provided here, feel free to reach out on twitter ([@vykthur](https://twitter.com/vykthur)) and share your work! ## Motivation - Why Track/Detect hands with Neural Networks? There are several existing approaches to tracking hands in the computer vision domain. Incidentally, many of these approaches are rule based (e.g extracting background based on texture and boundary features, distinguishing between hands and background using color histograms and HOG classifiers,) making them not very robust. For example, these algorithms might get confused if the background is unusual or in situations where sharp changes in lighting conditions cause sharp changes in skin color or the tracked object becomes occluded.(see [here for a review](https://www.cse.unr.edu/~bebis/handposerev.pdf) paper on hand pose estimation from the HCI perspective) With sufficiently large datasets, neural networks provide opportunity to train models that perform well and address challenges of existing object tracking/detection algorithms - varied/poor lighting, noisy environments, diverse viewpoints and even occlusion. The main drawbacks to usage for real-time tracking/detection is that they can be complex, are relatively slow compared to tracking-only algorithms and it can be quite expensive to assemble a good dataset. But things are changing with advances in fast neural networks. Furthermore, this entire area of work has been made more approachable by deep learning frameworks (such as the tensorflow object detection api) that simplify the process of training a model for custom object detection. More importantly, the advent of fast neural network models like ssd, faster r-cnn, rfcn (see [here](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md#coco-trained-models-coco-models) ) etc make neural networks an attractive candidate for real-time detection (and tracking) applications. Hopefully, this repo demonstrates this. > If you are not interested in the process of training the detector, you can skip straight to applying the [pretrained model I provide in detecting hands](#detecting-hands). Training a model is a multi-stage process (assembling dataset, cleaning, splitting into training/test partitions and generating an inference graph). While I lightly touch on the details of these parts, there are a few other tutorials cover training a custom object detector using the tensorflow object detection api in more detail[ see [here](https://pythonprogramming.net/training-custom-objects-tensorflow-object-detection-api-tutorial/) and [here](https://towardsdatascience.com/how-to-train-your-own-object-detector-with-tensorflows-object-detector-api-bec72ecfe1d9) ]. I recommend you walk through those if interested in training a custom object detector from scratch. ## Data preparation and network training in Tensorflow (Dataset, Import, Training) **The Egohands Dataset** The hand detector model is built using data from the [Egohands Dataset](http://vision.soic.indiana.edu/projects/egohands/) dataset. This dataset works well for several reasons. It contains high quality, pixel level annotations (>15000 ground truth labels) where hands are located across 4800 images. All images are captured from an egocentric view (Google glass) across 48 different environments (indoor, outdoor) and activities (playing cards, chess, jenga, solving puzzles etc). <img src="images/egohandstrain.jpg" width="100%"> If you will be using the Egohands dataset, you can cite them as follows: > Bambach, Sven, et al. "Lending a hand: Detecting hands and recognizing activities in complex egocentric interactions." Proceedings of the IEEE International Conference on Computer Vision. 2015. The Egohands dataset (zip file with labelled data) contains 48 folders of locations where video data was collected (100 images per folder). ``` -- LOCATION_X -- frame_1.jpg -- frame_2.jpg ... -- frame_100.jpg -- polygons.mat // contains annotations for all 100 images in current folder -- LOCATION_Y -- frame_1.jpg -- frame_2.jpg ... -- frame_100.jpg -- polygons.mat // contains annotations for all 100 images in current folder ``` **Converting data to Tensorflow Format** Some initial work needs to be done to the Egohands dataset to transform it into the format (`tfrecord`) which Tensorflow needs to train a model. This repo contains `egohands_dataset_clean.py` a script that will help you generate these csv files. - Downloads the egohands datasets - Renames all files to include their directory names to ensure each filename is unique - Splits the dataset into train (80%), test (10%) and eval (10%) folders. - Reads in `polygons.mat` for each folder, generates bounding boxes and visualizes them to ensure correctness (see image above). - Once the script is done running, you should have an images folder containing three folders - train, test and eval. Each of these folders should also contain a csv label document each - `train_labels.csv`, `test_labels.csv` that can be used to generate `tfrecords` Note: While the egohands dataset provides four separate labels for hands (own left, own right, other left, and other right), for my purpose, I am only interested in the general `hand` class and label all training data as `hand`. You can modify the data prep script to generate `tfrecords` that support 4 labels. Next: convert your dataset + csv files to tfrecords. A helpful guide on this can be found [here](https://pythonprogramming.net/creating-tfrecord-files-tensorflow-object-detection-api-tutorial/).For each folder, you should be able to generate `train.record`, `test.record` required in the training process. ## Training the hand detection Model Now that the dataset has been assembled (and your tfrecords), the next task is to train a model based on this. With neural networks, it is possible to use a process called [transfer learning](https://www.tensorflow.org/tutorials/image_retraining) to shorten the amount of time needed to train the entire model. This means we can take an existing model (that has been trained well on a related domain (here image classification) and retrain its final layer(s) to detect hands for us. Sweet!. Given that neural networks sometimes have thousands or millions of parameters that can take weeks or months to train, transfer learning helps shorten training time to possibly hours. Tensorflow does offer a few models (in the tensorflow [model zoo](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md#coco-trained-models-coco-models)) and I chose to use the `ssd_mobilenet_v1_coco` model as my start point given it is currently (one of) the fastest models (read the SSD research [paper here](https://arxiv.org/pdf/1512.02325.pdf)). The training process can be done locally on your CPU machine which may take a while or better on a (cloud) GPU machine (which is what I did). For reference, training on my macbook pro (tensorflow compiled from source to take advantage of the mac's cpu architecture) the maximum speed I got was 5 seconds per step as opposed to the ~0.5 seconds per step I got with a GPU. For reference it would take about 12 days to run 200k steps on my mac (i7, 2.5GHz, 16GB) compared to ~5hrs on a GPU. > **Training on your own images**: Please use the [guide provided by Harrison from pythonprogramming](https://pythonprogramming.net/training-custom-objects-tensorflow-object-detection-api-tutorial/) on how to generate tfrecords given your label csv files and your images. The guide also covers how to start the training process if training locally. [see [here] (https://pythonprogramming.net/training-custom-objects-tensorflow-object-detection-api-tutorial/)]. If training in the cloud using a service like GCP, see the [guide here](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/running_on_cloud.md). As the training process progresses, the expectation is that total loss (errors) gets reduced to its possible minimum (about a value of 1 or thereabout). By observing the tensorboard graphs for total loss(see image below), it should be possible to get an idea of when the training process is complete (total loss does not decrease with further iterations/steps). I ran my training job for 200k steps (took about 5 hours) and stopped at a total Loss (errors) value of 2.575.(In retrospect, I could have stopped the training at about 50k steps and gotten a similar total loss value). With tensorflow, you can also run an evaluation concurrently that assesses your model to see how well it performs on the test data. A commonly used metric for performance is mean average precision (mAP) which is single number used to summarize the area under the precision-recall curve. mAP is a measure of how well the model generates a bounding box that has at least a 50% overlap with the ground truth bounding box in our test dataset. For the hand detector trained here, the mAP value was **0.9686@0.5IOU**. mAP values range from 0-1, the higher the better. <img src="images/accuracy.jpg" width="100%"> Once training is completed, the trained inference graph (`frozen_inference_graph.pb`) is then exported (see the earlier referenced guides for how to do this) and saved in the `hand_inference_graph` folder. Now its time to do some interesting detection. ## Using the Detector to Detect/Track hands If you have not done this yet, please following the guide on installing [Tensorflow and the Tensorflow object detection api](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md). This will walk you through setting up the tensorflow framework, cloning the tensorflow github repo and a guide on - Load the `frozen_inference_graph.pb` trained on the hands dataset as well as the corresponding label map. In this repo, this is done in the `utils/detector_utils.py` script by the `load_inference_graph` method. ```python detection_graph = tf.Graph() with detection_graph.as_default(): od_graph_def = tf.GraphDef() with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid: serialized_graph = fid.read() od_graph_def.ParseFromString(serialized_graph) tf.import_graph_def(od_graph_def, name='') sess = tf.Session(graph=detection_graph) print("> ====== Hand Inference graph loaded.") ``` - Detect hands. In this repo, this is done in the `utils/detector_utils.py` script by the `detect_objects` method. ```python (boxes, scores, classes, num) = sess.run( [detection_boxes, detection_scores, detection_classes, num_detections], feed_dict={image_tensor: image_np_expanded}) ``` - Visualize detected bounding detection_boxes. In this repo, this is done in the `utils/detector_utils.py` script by the `draw_box_on_image` method. This repo contains two scripts that tie all these steps together. - detect_multi_threaded.py : A threaded implementation for reading camera video input detection and detecting. Takes a set of command line flags to set parameters such as `--display` (visualize detections), image parameters `--width` and `--height`, videe `--source` (0 for camera) etc. - detect_single_threaded.py : Same as above, but single threaded. This script works for video files by setting the video source parameter videe `--source` (path to a video file). ```cmd # load and run detection on video at path "videos/chess.mov" python detect_single_threaded.py --source videos/chess.mov ``` > Update: If you do have errors loading the frozen inference graph in this repo, feel free to generate a new graph that fits your TF version from the model-checkpoint in this repo. Use the [export_inference_graph.py](https://github.com/tensorflow/models/blob/master/research/object_detection/export_inference_graph.py) script provided in the tensorflow object detection api repo. More guidance on this [here](https://pythonprogramming.net/testing-custom-object-detector-tensorflow-object-detection-api-tutorial/?completed=/training-custom-objects-tensorflow-object-detection-api-tutorial/). ## Thoughts on Optimization. A few things that led to noticeable performance increases. - Threading: Turns out that reading images from a webcam is a heavy I/O event and if run on the main application thread can slow down the program. I implemented some good ideas from [Adrian Rosebuck](https://www.pyimagesearch.com/2017/02/06/faster-video-file-fps-with-cv2-videocapture-and-opencv/) on parrallelizing image capture across multiple worker threads. This mostly led to an FPS increase of about 5 points. - For those new to Opencv, images from the `cv2.read()` method return images in [BGR format](https://www.learnopencv.com/why-does-opencv-use-bgr-color-format/). Ensure you convert to RGB before detection (accuracy will be much reduced if you dont). ```python cv2.cvtColor(image_np, cv2.COLOR_BGR2RGB) ``` - Keeping your input image small will increase fps without any significant accuracy drop.(I used about 320 x 240 compared to the 1280 x 720 which my webcam provides). - Model Quantization. Moving from the current 32 bit to 8 bit can achieve up to 4x reduction in memory required to load and store models. One way to further speed up this model is to explore the use of [8-bit fixed point quantization](https://heartbeat.fritz.ai/8-bit-quantization-and-tensorflow-lite-speeding-up-mobile-inference-with-low-precision-a882dfcafbbd). Performance can also be increased by a clever combination of tracking algorithms with the already decent detection and this is something I am still experimenting with. Have ideas for optimizing better, please share! <img src="images/general.jpg" width="100%"> Note: The detector does reflect some limitations associated with the training set. This includes non-egocentric viewpoints, very noisy backgrounds (e.g in a sea of hands) and sometimes skin tone. There is opportunity to improve these with additional data. ## Integrating Multiple DNNs. One way to make things more interesting is to integrate our new knowledge of where "hands" are with other detectors trained to recognize other objects. Unfortunately, while our hand detector can in fact detect hands, it cannot detect other objects (a factor or how it is trained). To create a detector that classifies multiple different objects would mean a long involved process of assembling datasets for each class and a lengthy training process. > Given the above, a potential strategy is to explore structures that allow us **efficiently** interleave output form multiple pretrained models for various object classes and have them detect multiple objects on a single image. An example of this is with my primary use case where I am interested in understanding the position of objects on a table with respect to hands on same table. I am currently doing some work on a threaded application that loads multiple detectors and outputs bounding boxes on a single image. More on this soon.

Computer Vision Projects with OpenCV and Python 3, published by Packt

A computer vision project tracking an object in space with a camera actuated by two stepper motors. Uses an Arduino microcontroller for stepper motor control, and the Python 3 OpenCV library for computer vision.

Actual Project file https://drive.google.com/file/d/1RCJ271K1B5Ig839c_0UCq8oWn5mpz7EN/view?usp=sharing Introduction This project was part of the embedded system design course, and uses face recognition to control a servo lock. The face recognition has been done using the Eigenfaces algorithm (Principle Component Analysis or PCA) and implemented using the Python API of OpenCV. Open Source Project source It's a slight modification of the Raspberry Pi Face Recognition Treasure Box project by Tony Dicola on the Adafruit Learning System. The code has been modified at places to replace the use of the RPIO library (which has issues running on the new Raspberry Pi 2 Model B+) with the standard RPi.GPIO library. The project has also been implemented to work as an automated home lock system which unlocks for the owner of the house and doesn't for any other visitor. It also plays an appropriate voice message. IMPLEMENTATION DETAILS This slight modification also changed the way of installing the dependencies,OpenCV & Python version and also the installation of updated GPIO ports for Raspberry B+. The modifications that has done here also includes the .wave sound files that tends to start or stop depending upon the door recognition status. OpenCV Installation This project depends on the OpenCV computer vision library to perform the face detection and recognition. Unfortunately the current binary version of OpenCV available to install in the Raspbian operating system through apt-get (version 2.3.x) is too old to contain the face recognition algorithms used by this project. However you can download, compile, and install a later version of OpenCV to access the face recognition algorithms. Note: Compiling OpenCV on the Raspberry Pi will take about 3 hours of mostly unattended time. Make sure you have some time to start the process before proceeding. First you will need to install OpenCV dependencies before you can compile the code. Connect to your Raspberry Pi in a terminal session and execute the following command: sudo apt-get update sudo apt-get install build-essential cmake pkg-config python-dev libgtk2.0-dev libgtk2.0 zlib1g-dev libpng-dev libjpeg-dev libtiff-dev libjasper-dev libavcodec-dev swig unzip Answer yes to any questions about proceeding and wait for the libraries and dependencies to be installed. You can ignore messages about packages which are already installed. Next you should download and unpack the OpenCV source code by executing the following commands: wget http://downloads.sourceforge.net/project/opencvlibrary/opencv-unix/2.4.10/opencv-2.4.10.zip unzip opencv-2.4.10.zip Note that this project was written using OpenCV 2.4.10, although any 2.4.x version of OpenCV should have the necessary face recognition algorithms. Now change to the directory with the OpenCV source and execute the following cmake command to build the makefile for the project. Note that some of the parameters passed in to the cmake command will disable compiling performance tests and GPU accelerated algorithms in OpenCV. I found removing these from the OpenCV build was necessary to help reduce the compilation time, and successfully compile the project with the low memory available to the Raspberry Pi. cd opencv-2.4.9 cmake -DCMAKE_BUILD_TYPE=RELEASE -DCMAKE_INSTALL_PREFIX=/usr/local -DBUILD_PERF_TESTS=OFF -DBUILD_opencv_gpu=OFF -DBUILD_opencv_ocl=OFF After this command executes you should see details about the build environment and finally a '-- Build files have been written to: ...' message. You might see a warning that the source directory is the same as the binary directory--this warning can be ignored (most cmake projects build inside a subdirectory of the source, but for some reason I couldn't get this to work with OpenCV and built it inside the source directory instead). If you see any other error or warning, make sure the dependencies above were installed and try executing the cmake command again. Next, compile the project by executing: make This process will take a significant amount of time (about 3 hours), but you can leave it unattended as the code compiles. Finally, once compilation is complete you can install the compiled OpenCV libraries by executing: sudo make install After this step the latest version of OpenCV should be installed on your Raspberry Pi. Python Dependencies The code for this project is written in python and has a few dependencies that must be installed. Once connected to your Raspberry Pi in a terminal session, execute the following commands: sudo apt-get install python-pip sudo apt-get install python-dev sudo pip install picamera sudo pip install RPi.GPIO You can ignore any messages about packages which are already installed or up to date. These commands will install the picamera library for access to the Raspberry Pi camera, and the GPIO library for access to the Pi GPIO pins and PWM support. Hardware The Hardware required for this project are as follows: Raspberry Pi ( I prefer Model 2 B+) Raspberry Pi Camera Micro Servo One Push Button Power Supply for the Servo (5V Source) One 10K resistor for pull down Breadboard and Jumper wires for connections The necessary circuit diagrams and further explanations are explained in depth in the original pdf accompanying the project. Kindly go through it first.

During winter break of 2016, I self taught myself openCV and various Computer Vision concepts. I learn best by doing projects so this repo contains all of my practices and mini projects with openCV 3.0 and python 3.5.

HackDavis 2025 project. VIVI empowers neurodivergent children to express imagination through gaze-activated AI and tracks reading stats in an all in one platform. Built with Python/FastAPI, OpenCV computer vision, OpenAI Whisper, DALL-E 3, and ElevenLabs voice agents.

In this project,a system has been developed that finds the plate on the vehicle with image processing algorithms.This systems uses on security sistems,common etrance areas,car park entrances and exits,traffic control,university entrances and exits,building entrances and exits.Through this system aimed to minimize human power,cost and security threats.This project developed by Python programming language and OpenCV 3.3 (Open Source computer vision library).Experiments were performed on images taken from different locations and taken from different media.

With the advancement of modern technologies areas related to computer vision and real time image processing has become a major technology under consideration. Our aim in this project is to combine the processes of object detection, colour detection and object tracking to implement what can be used as a virtual drawing board. Our proposed software detects objects from the webcam in real-time and tracks their movement to replicate the object’s path of motion as a drawing on the screen. A colour detection module is embedded which to allow only objects of a particular colour (blue here) to be used as the painting stick. All this will be achieved using the OpenCV library of Python. By using open source computer vision library (OpenCV for short), an image can be captured on the bases of its hue, saturation and colour value (HSV) range. The basic library functions for image handling and processing are used. The new features which we will attempt to embed in this project are the following; 1. Touch-less interface 2. Any object of any colour and size can be used 3. Automatic colour detection and selection based on the object's colour 4. Infant-friendly application which can be used to teach colours in fun way 5. Interactive and engaging interface for autistic patients

Study from book: Computer Vision Projects with OpenCV and Python 3

Computer Vision project. Detects different coins from the Mario game with assigned values and sums the result with MAE < 3.0. Written using python with OpenCV

Dockerized template for new projects of computer vision on python 3.11 with cuda 12.4 and a lot of libraries installed (opencv, torch, ultralytics, pillow, albumentations...)

A real-time face detection project using OpenCV. Captures video from a webcam, detects faces using the Haar Cascade model, and highlights them with green rectangles. Easy to set up and run with Python 3.x and OpenCV. Ideal for beginners exploring computer vision. Press `ESC` to exit.

The Research Project is in the field of Computer Vision. It was created for solving problems with Facial Recognition with a image with a lot of faces and recognize one by one. All the Project was developed with Python 3 using the OpenCV librarie and ORB Framework. The project uses SADE Algorithm (Self-adaptive Differential Evolution) as a computacional intelligence algorithm.