Video Call Terminal Inappropriate Content Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users making video calls may be exposed to inappropriate content, leading to discomfort or shame, as existing technologies lack effective measures to filter out obscene videos in real-time during video calls.
Innovation Solution
A terminal with a processor that detects specific events, such as terminal inclination or disappearance of a user's face, uses a classification model to determine if inappropriate content is present, and stops transmitting the video or sound accordingly, while displaying alternative content if necessary, and resumes transmission when the content is deemed appropriate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video transmission is continuously monitored and filtered in real-time during video calls, then user protection from inappropriate content is improved, but device complexity and processing overhead increase
Solution Approach 1:
The system performs preliminary classification model training offline to create pre-computed decision boundaries and filtering rules. During real-time video calls, the pre-trained model enables rapid inference without requiring complex online computation, thus protecting users from inappropriate content while minimizing processing overhead during the actual video call.
Solution Approach 2:
The patent introduces an intermediary classification model that acts as a mediator between the video transmission source and the receiver. This model filters and classifies video content before transmission or during transmission, preventing inappropriate content from reaching users without requiring complex end-to-end verification systems.
2Object-affected harmful factors
If video transmission is stopped when inappropriate content is detected, then user exposure to inappropriate content is prevented, but loss of useful video information occurs
Solution Approach 1:
The system extracts and removes only the inappropriate portions or segments of video content while allowing the rest of the video transmission to continue. This selective extraction approach prevents exposure to harmful content without unnecessarily interrupting the overall video call or losing valuable video information.
Solution Approach 2:
The classification model applies different quality filters to different regions or segments of the video stream. Inappropriate regions are blocked or blurred while appropriate regions continue to transmit normally, maintaining local quality control rather than applying a global stop-all approach to video transmission.
3Measurement precision
If classification model processing is performed on every video frame, then detection accuracy is improved, but processing speed and energy consumption decrease
Solution Approach 1:
Instead of processing every video frame, the system implements periodic sampling where the classification model is applied at regular intervals or at key frames only. This periodic processing maintains adequate detection accuracy for identifying inappropriate content while significantly improving processing speed and reducing energy consumption compared to frame-by-frame analysis.
Solution Approach 2:
The system performs classification processing on a partial subset of video data (e.g., key frames, representative samples) rather than the complete video stream. This partial action approach provides sufficient detection capability to identify inappropriate content while avoiding the excessive processing burden of analyzing every single frame.
Data Source
AI summary
Provided is a terminal. The terminal may include an input and output interface configured to receive a first video capturing a first user, a communication interface configured to transmit the first video or a first sound of the first user to a terminal of a counterpart and receive a second video or second sound, which is obtained by the terminal of the counterpart, through a video call session, a display configured to display the first video and the second video, and a processor configured to detect a specific event during the video call session, determine whether an inappropriate element is included in the first video or the first sound when the specific event is detected, and stop transmitting the first video or the first sound according to a result of the determination.