Add moderation step for images

[e-caste]

The current pipeline is purely focused on language-based moderation, meaning that it won't care about any image (which we will filter out based on our custom enhanced MarkDown format, ref #97).

The base64-encoded images provided by the users should also be checked for NSFW content. Find a pre-trained model for this task, we don't need to build a dataset of our own for this objective.

Also to note is that users could upload large/very large images, so the model (likely CNN-based) should automatically rescale them to a resolution which allows for a reasonable inference time.

This could be a starting point: https://github.com/SashiDo/content-moderation-image-api

---

• [ ] find some pre-trained models and test them for our purpose -- keeping the same dataset (e.g. the one from nsfwjs.com) test a few models against it and compare the accuracy and the inference time on the same platform (both CPU and GPU, e.g. nsfwjs.com runs with Tensorflow.js on GPU accelerated by WebGL but in <1 second so on CPU it probably doesn't take a lot more)
• [ ] support 2 image input kinds: URL (from the pure MarkDown syntax) and base64 encoded (contained within our enhanced MarkDown format) -- you can check out what a base64-encoded image looks like by running base64 --input <path to image>
• [ ] pre-processing steps: rescale image, color fix-up -- maximum size should be determined by the model's input specifications
• [ ] detect text contained inside the image (e.g. memes, probably with OCR) -- the detected text should be forwarded to the NLP text pipeline
• [ ] define a "toxicity" policy (no nudes, porn, politics, insults, other stuff which shouldn't be on an e-learning platform)
• [ ] what about gifs? should we run the model against each frame, disallow them entirely, run the model against a few frames..?

[e-caste]

About point 1:
the nsfw-detector library (Tensorflow-based) seems to work quite well with random images found online (TODO: test it against this dataset (just run the Docker commands docker build . -t docker_nsfw_data_scraper and docker run -v $(pwd):/root/nsfw_data_scraper docker_nsfw_data_scraper scripts/runall.sh): https://github.com/alex000kim/nsfw_data_scraper)

See the test notebook here: https://colab.research.google.com/drive/1y_w0t1ncJwN3vt4xNt0s_EIK_EHggO43?usp=sharing And the test data on Google Drive here: https://drive.google.com/drive/folders/1r_FKqwFpVnr29CtYYJAYTo0txAnjFNVy?usp=sharing

[FereshtehFeiz]

working on the point 4 for extracting the text on the images and I found the below python script with OCR that detects the text in save it in txt file. https://www.geeksforgeeks.org/text-detection-and-extraction-using-opencv-and-ocr/

[e-caste]

Hugging Face just released huggingface.js, so this step could be delegated to their APIs entirely directly from the browser (must clarify API usage cost). https://github.com/huggingface/huggingface.js/blob/main/packages/inference/README.md