Embedded Files
Problem Statment
In today's society, everything moves quickly, and no one wants to waste time watching meaningless stuff. Manually examining CCTV footage, watching whole matches rather than highlights, and listening to lengthy video lectures are all time-consuming tasks in surveillance. There is no effective method to avoid it. Additionally, privacy concerns arise when sensitive information like as faces and vehicle plates are shown in raw footages. Large video files might be challenging to store and share. There is a rising demand for an automated, intelligent system that can not only summarize lengthy movies into crucial events, but also safeguard privacy through masking and assure effective storage through smart compression—all without relying heavily on human work.
Existing System/ Related Works
Proposed System Modules
1. User Authentication: Secure user login and registration using Gmail or Google Account.
2. Video Upload & Management: Allow user to upload video for preprocessing. User can trim specific part of video in order to apply summarization on it. User can preview his previous videos in order to manage them.
3. Video Summarization: This module will extract the key moments from video based on user query and model training. It will convert video into short fragmentations and score them. After choosing based on score and user query, it will generate a shorter video.
4. Video Summarization in Text: It will explain the video in text format. It will help users to read video summary timestamp wise.
5. Sensitive Information blurring: It will detect faces, ID cards and vehicle’s number plates to blur them in order to maintain the privacy.
6. Video Compression: It will compress size the video size into half without losing most of video quality. So, that it will be easy to share them.
Advantage of the Proposed System
Efficient query-based video compression with less reliance on manual effort:
Efficient query-based video compression with less reliance on manual effort:
This project introduces a smart video summarization system that identifies and retains only the most relevant parts of a video based on automated analysis or user queries and remove irrelevant part of the video. By minimizing manual intervention, the system significantly reduces the time and effort required for editing, making it easier to extract meaningful content without losing context.
Enhanced privacy through precise and automated detection and masking of sensitive information:
To ensure data protection and user confidentiality, the system integrates advanced AI models that automatically detect and blur sensitive elements such as faces, personal
IDs, or confidential visuals in videos. This approach minimizes human error and guarantees consistent privacy measures across large video datasets.
Optimized file sizes to ease storage demands and improve transfer speeds:
Optimized file sizes to ease storage demands and improve transfer speeds:
By intelligently compressing video files without compromising essential visual information, the system drastically reduces storage requirements. This not only saves physical storage space but also accelerates uploading, downloading, and streaming processes, making it ideal for environments with limited bandwidth or storage capabilities.
Synopsis Document
NeuroClip_FYP_Proposal.pdfSynopsis Presentation
NeuroClip_Presentation.pptxApp Screenshots
Team
Page updated
Google Sites
Report abuse