Detecting Faces

[abhisheks008]

Deep Learning Simplified Repository (Proposing new issue)

:red_circle: Project Title : Detecting Faces :red_circle: Aim : The aim is to create a deep learning project which will detect the faces using OpenCV and MTCNN approach with an accuracy over 85%. :red_circle: Dataset : https://www.kaggle.com/code/wittmannf/detecting-faces-using-opencv-mtcnn-no-internet/data :red_circle: Approach : Try to use 3-4 algorithms to implement the models and compare all the algorithms to find out the best fitted algorithm for the model by checking the accuracy scores. Also do not forget to do a exploratory data analysis before creating any model.

---

📍 Follow the Guidelines to Contribute in the Project :

• You need to create a separate folder named as the Project Title.
• Inside that folder, there will be four main components.
  • Images - To store the required images.
  • Dataset - To store the dataset or, information/source about the dataset.
  • Model - To store the machine learning model you've created using the dataset.
  • requirements.txt - This file will contain the required packages/libraries to run the project in other machines.
• Inside the Model folder, the README.md file must be filled up properly, with proper visualizations and conclusions.

---

:red_circle::yellow_circle: Points to Note :

• The issues will be assigned on a first come first serve basis, 1 Issue == 1 PR.
• "Issue Title" and "PR Title should be the same. Include issue number along with it.
• Follow Contributing Guidelines & Code of Conduct before start Contributing.

---

:white_check_mark: To be Mentioned while taking the issue :

• Full name :
• GitHub Profile Link :
• Email ID :
• Participant ID (if applicable):
• Approach for this Project :
• What is your participant role? (Mention the Open Source program)

---

Happy Contributing 🚀

All the best. Enjoy your open source journey ahead. 😎

[AbhishekRP2002]

Hey there ! Can I work on this issue ? I would like to contribute to this .

[abhisheks008]

✅ To be Mentioned while taking the issue :

Full name : GitHub Profile Link : Email ID : Participant ID (if applicable): Approach for this Project : What is your participant role? (Mention the Open Source program)

Please mention the credentials. @AbhishekRP2002

[AbhishekRP2002]

Full name : Abhishek Ranjan Prusty GitHub Profile Link : https://github.com/AbhishekRP2002 Email ID : [email protected] Participant ID (if applicable): Approach for this Project : This model that we will use for Face detection will be Multi-task Cascaded Convolutional Networks (MTCNN), which is essentially several convolutional networks strung together that give out several pieces of information. The neural network is expected to detect individual faces, and locate facial landmarks (i.e. two eyes, nose, and endpoints of the mouth), along with drawing a bounding box around the face. We will use OpenCV to open, read, write, and show images. What is your participant role? (SSOC 2022)

[Sahana915]

Issue no: #8 Detecting Faces Full Name : Sahana B Kanteppagoudra Github profile link: https://github.com/Sahana915 Participation role : SSOC 2022 Approach: I will try out by using variety of DL algorithms like Haar Cascade or I'll go by using HOG based detector in dlib .

[Rushijaviya]

Assigned to @AbhishekRP2002. @Sahana915, you can choose other issues; we assign issues based on FCFS.

[abhisheks008]

I am sorry to say that @AbhishekRP2002 has already assigned an issue, complete that one first. This issue is assigned to @Sahana915.

@AbhishekRP2002 Do not comment out in other issues while you have already assigned to an issue. Finish that first, then take other issues.

[abhisheks008]

@Sahana915 updates please.

[abhisheks008]

Updates please @Sahana915

[Singhal-Akshat]

Full name: Akshat Singhal GitHub Profile Link : https://github.com/Singhal-Akshat Email ID: [email protected] Approach for this Project: I will use MTCNN and OpenCV, along with a comparison with the Haarcascade method, for MTCNN will use Dlib. What is your participant role? SSOC (2023)

[abhisheks008]

Issue assigned to you @Singhal-Akshat

[saikrishna823]

hi, @abhisheks008 ,could you please assign issue to me. Full Name:Mule Sai Krishna Reddy Github Profile Link:https://github.com/saikrishna823 Email ID:[email protected] Participant ID (if applicable): Approach for this Project : I will use MTCNN,Haar cascade opencv ,yolo,SSD too to detect faces in real time. What is your participant role? I am participating as contributor through gssoc' 24.

[abhisheks008]

Hi @saikrishna823 thanks for showing up with your approach. If you can implement these algorithms this will be good project to include in this project repo. Stick to your approach and make sure you implement as many algorithms you can.

Assigned this issue to you @saikrishna823

[saikrishna823]

@abhisheks008 I am facing issues with downloading dataset.

[abhisheks008]

What are the issues you are facing?

[saikrishna823]

Not able to download dataset.It is under competiton.

[abhisheks008]

Ohhh! Can you find another dataset of same use from a trusted source (Kaggle will be better) for this project?

[saikrishna823]

okay

[Diyaa0313]

Full name : Diya Sen GitHub Profile Link : https://github.com/Diyaa0313 Email ID : [email protected] Participant ID (if applicable): NA Approach for this Project :

1. I will use EDA to plot the distribution of faces per image and use histograms to understand the variation in face sizes and orientation and remove any imbalance with oversampling or undersampling.

2. Use MTCNN, SSD, and YOLO for face detection and optimize the hyperparameters.

3. Compare which model gives higher accuracy.

4. Run the model on the image to detect faces. Draw bounding boxes around the detected faces using OpenCV.

What is your participant role? GSSOC 2024 Contributor

[abhisheks008]

Full name : Diya Sen GitHub Profile Link : https://github.com/Diyaa0313 Email ID : [email protected] Participant ID (if applicable): NA Approach for this Project :

1. I will use EDA to plot the distribution of faces per image and use histograms to understand the variation in face sizes and orientation and remove any imbalance with oversampling or undersampling.
2. Use MTCNN, SSD, and YOLO for face detection and optimize the hyperparameters.
3. Compare which model gives higher accuracy.
4. Run the model on the image to detect faces. Draw bounding boxes around the detected faces using OpenCV.

What is your participant role? GSSOC 2024 Contributor

This issue is already assigned to a contributor. You can check out the issues which are having Up for Grabs tag/label,

[saikrishna823]

Hi, @abhisheks008 I found image dataset.If it is okay for you,I will start implementing the project

[abhisheks008]

Hi, @abhisheks008 I found image dataset.If it is okay for you,I will start implementing the project

Can you please share the source/URL of the dataset before start working on it? @saikrishna823

[saikrishna823]

Hi @abhisheks008 ,I am sharing the url of the dataset http://shuoyang1213.me/WIDERFACE/

[abhisheks008]

Cool you can go ahead with this issue. Assigning this issue to you @saikrishna823

[why-Radhika]

I noticed it has been more than a month since the issue has been assigned and there is no PR for the same. So, I would like to work on this issue.

Full name : Radhika Gupta GitHub Profile Link : https://github.com/why-Radhika Email ID : [email protected] Approach for this Project : Multi-task Cascaded Convolutional Networks (MTCNN). MTCNN consists of multiple convolutional networks arranged in a cascaded manner, designed to output several pieces of information simultaneously. Specifically, it will detect individual faces within images, accurately locate facial landmarks such as the eyes, nose, and endpoints of the mouth, and draw bounding boxes around detected faces. In conjunction with MTCNN, we will leverage OpenCV for tasks such as image manipulation, reading and writing images, and visualizing the detection results. What is your participant role? (Mention the Open Source program) : GSSOC'24

[abhisheks008]

Hi @why-Radhika go ahead with the issue.

Unassigning this issue @saikrishna823 due to inactivity.