This project aims to develop a computer vision solution that seamlessly inserts an advertisement image into a video, addressing occlusions caused by various objects or movements within the scene.

The primary objective is to integrate the advertisement image into the video content naturally, ensuring it blends well and gracefully handles occlusions.

The scope involves creating an algorithm that effectively places the advertisement image into the video, ensuring it integrates well with the content and appropriately addresses occlusions. The focus is on achieving a natural and unobtrusive insertion despite potential occlusions in the scene.

OpenCV: Used for image processing and manipulation.

Python (version 3.x): Chosen programming language for its extensive support in computer vision tasks.

XVID Codec: Used for encoding the output video file.

Other standard Python libraries for general-purpose programming and file handling.

The algorithm follows a structured approach:

Background Removal: The advertisement image undergoes background removal to ensure transparency, saving it with an alpha channel.

ROI Detection: Automatic detection of the Region of Interest (ROI) within sample video frames using Canny edge detection and contour analysis.

Scaling Factors: Calculation of scaling factors to adjust ROI coordinates based on resolution differences between sample and input videos.

Occlusion Handling: Implementation of strategic occlusion handling techniques, blending the advertisement image seamlessly with occluded regions using alpha blending.

Video Output: Processed frames are compiled into an output video file ("output_video.mp4") using the XVID codec, demonstrating successful advertisement integration with occlusion handling.

The developed algorithm successfully meets its objectives by seamlessly integrating the advertisement image into the video while effectively handling occlusions. Strategic methodologies and insights gained during development contribute to the robustness and visual appeal of the solution.