Complete Implementation for Disparity Map Estimation

In this project, we will develop code for complete disparity map estimation from stereo image pairs consisting of right and left images. We will employ computer vision techniques and algorithms to achieve this objective. A disparity map represents an image format that facilitates depth perception and 3D reconstruction. By estimating the disparity between corresponding points in right and left images, we can extract depth information for various objects within the scene. This capability is crucial for numerous applications including robotic navigation, stereoscopic imaging, and virtual reality systems. Our implementation will involve multiple computational stages: image preprocessing, feature extraction, matching algorithms, and disparity calculation. For image preprocessing, we may utilize Gaussian filtering to reduce noise while preserving edge information. Feature extraction could employ SIFT (Scale-Invariant Feature Transform) or ORB (Oriented FAST and Rotated BRIEF) detectors to identify keypoints in both images. The matching phase will implement algorithms like Block Matching or Semi-Global Block Matching (SGBM) to establish correspondences between left and right image features. We will leverage established computer vision libraries such as OpenCV to streamline development, ensuring our code produces accurate disparity maps. The implementation will include post-processing steps like left-right consistency checks and weighted least squares filtering to refine the final disparity output. The code structure will modularize each processing stage, allowing for parameter optimization and algorithm comparisons.