In-Frame Video Markers for Ad Insertion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Small content providers face high costs and complexity in inserting advertisements into streamed video content, as they require specialized hardware and software to embed ad markers within video feeds.
Innovation Solution
The streaming platform detects recorded objects within video frames as content triggers to insert other video content, such as advertisements, reducing the need for costly equipment by using these objects as markers for determining the timing and position of ad insertion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If specialized hardware devices and software are used to insert advertisement markers within video feeds, then advertisement insertion capability is achieved, but cost and device complexity increase significantly
Solution Approach 1:
The video content itself provides the markers needed for advertisement insertion through object detection and recognition. The system automatically identifies objects within video frames and uses them as triggers, eliminating the need for external specialized hardware to embed markers. This self-service approach allows content to be inherently ad-ready without additional equipment.
Solution Approach 2:
The patent replaces the mechanical hardware system (specialized hardware devices for marker insertion) with a software-based computer vision system. Instead of physically embedding markers through hardware, the system uses software to detect and recognize objects within video frames, substituting mechanical operations with computational processing.
2Reliability
If specialized hardware devices are used to insert advertisement markers, then advertisement functionality is enabled, but cost increases significantly for small content providers
Solution Approach 1:
The system replaces expensive, permanent specialized hardware with inexpensive software-based object detection. Instead of investing in costly hardware devices that remain fixed in the system, the patent uses flexible software algorithms that can be deployed on standard computing infrastructure, dramatically reducing costs for small content providers.
Solution Approach 2:
The patent uses computer vision to create digital copies or representations of objects within video frames. These detected objects serve as virtual markers that trigger advertisements, replacing the need for physical marker embedding hardware. This copying approach allows standard video content to function with advertisement capabilities without specialized equipment.
3Device complexity
If recorded objects within video frames are used as content triggers, then cost and complexity are reduced, but detection and measurement difficulty increases
Solution Approach 1:
The patent employs universal object detection algorithms that can identify multiple types of objects across diverse video content using the same system. The computer vision model is trained to recognize various objects (people, products, scenes) and can adapt to different contexts, making the detection process more straightforward despite the variety of potential targets.
Solution Approach 2:
The system incorporates feedback mechanisms where detected objects are continuously refined and validated. The object detection process uses feedback from frame analysis, confidence scoring, and pattern recognition to improve accuracy over time, making the detection process more reliable and easier to manage despite the complexity of analyzing video content.
Data Source
AI summary
Techniques are described for performing actions (e.g., streaming video content, such as advertising content) based on visual markers detected within a live video feed being streamed to one or more computing devices. First video content is streamed to the one or more computing devices. A presence of a content item within the first video content is detected at a first point in time during streaming of the first video content. An action associated with the content item is performed at a second point in time in response to detection of the presence of the content item in the first video content.