Images and words, together
Compare image and text branches, then test eight fusion strategies for harmful-meme classification.
Inside the project
Research showcaseOriginal artifact ↗
How it works.
Separate the modalities
Process images with a ResNet-based CNN and text with the project's classification branch.
From the original project.
Saved artifacts · click to inspect
Original artifact ↗
Original artifact ↗
Continue in the source.
Open the source notebook in Jupyter, Colab, or the environment described in the README. Data and model downloads may be required.
git clone https://github.com/eforus-overseer/Multimodal-Harmful-Content-Detection-Social-Media-Memes-.gitProject artifacts.
[Experiment]-Precision-Optimized Harmful Content-Detection.ipynb ↗[Exploration]-Harmful-Content-Detection-Texts-Analysis.ipynb ↗[Experiment]-Harmful-Meme-Text-LSTM.ipynb ↗Multimodal Harmful Content Detection in Social Media Memes.pdf ↗
Source links point to the original public repository. Credit belongs to the project authors and the dependencies credited there.