Emotion detection using CNN and transfer learning with a interactive flask webapp.

[Shantnu-singh]

Objective

a web app where you upload a picture and it detects the emotions on the faces! We can build this using a Convolutional Neural Network (CNN) trained to recognize emotions. To jumpstart the training process, we'll leverage transfer learning, where we adapt a pre-trained CNN model for image recognition to our specific emotion detection task. This web app, built with Flask, will provide an interactive way to understand and analyze emotions in pictures.

@apu52 Please Assign this Issue to Me under GSSOC24.

[github-actions[bot]]

Thanks for creating the issue,Please read the Pinned issued first and Readme.md in each Pull Request you made. Keep learning...

[invigorzz313]

@Shantnu-singh there is already a similar project in this repo.

[Shantnu-singh]

My idea is unique in the sense that it proposes a emotion detection tool that can track your emotions in real-time through an interactive web app. Unlike other similar idea in the repo, such as a music recommender based on emotion that uses OpenAI's API and requires users to manage their own API keys, my model will streamline the process. It will use CNN and pretrained models like VGG16 and ResNet50V2. This approach not only simplifies the user experience but also provides an excellent opportunity to understand how CNN models learn, how they can be fine-tuned, and how pretrained models like (VGG16 and ResNet50V2) can be adapted for specific use cases.

@invigorzz313 can you reconsider this on more time.

[invigorzz313]

@Shantnu-singh okay, go ahead