The problem NotesFrame solves

NotesFrame is an ML based Video to Text Notes maker that allows users to convert videos into text-based notes. The system utilizes advanced algorithms to extract key frames from the video and then generates text summaries from those frames, creating comprehensive notes for the video content in the pdf format.

Challenges we ran into

1.Extraction of Unnecessary Key Frames:

Initially, the model encountered challenges in extracting key frames from the video content.
The extracted key frames included unnecessary frames that did not contribute to the understanding or summarization of the video content.
Many of these frames lacked significant content or did not contain any relevant text, resulting in inefficient note generation.
Addressing this issue required refining the key frame extraction algorithm to ensure that only relevant frames, containing essential information, were selected.

2. Random Order Text Generation:

Another challenge encountered during the project was the generation of text in a random order.
The model for text extraction produced text snippets that were disorganized and did not follow a logical sequence.
This random arrangement of text hindered the coherence and readability of the generated notes, making it difficult for users to comprehend the content effectively.
To overcome this challenge, additional processing and sorting algorithms were implemented to organize the extracted text into a structured format, ensuring that the notes were presented in a logical and coherent manner.