Method for region of interest selection with noticeable stereoscopic distortions in S3D videos


Shooting 3D with two cameras without proper calibration causes geometric and sharpness distortions. The search of such distortions is manual and time-consuming. So, the special algorithm has been developed. It automates the process of fragments’ selection in stereoscopic frame containing the most noticeable geometric distortions and inconsistency of views in terms of sharpness.

Types of distortions

Types of distortions

Example of sharpness mismatch

Sharpness mismatch example

Example of color mismatch
Spy Kids 3D: Game Over

Color mismatch example

Example of rotation mismatch
Drive Angry

Rotation mismatch example Rotation mismatch example

Example of scale mismatch

Scale mismatch example

Example of vertical disparity
Journey to the Center of the Earth 3D

Vertical disparity example Vertical disparity example

Proposed method

Algorithm scheme

Algorithm scheme

The algorithms of region selection for frames, containing scale, rotation and/or sharpness mismatch, were improved through machine learning methods.

Machine learning


A dataset was created to train the model, which would predict the correctness of the detected region. The dataset was made by human experts who selected regions of interest, and consists of:

The results of classifiers (cross-validation and 95% confidence interval)

The results of classifiers C — regularization weight 𝛄 — kernel parameter

For each type of distortion, we chose the model that showed the best results. The model predicts the region that would likely be selected by an expert.


To decide whether the machine learning model is better than the baseline algorithm, we marked 100 additional frames and conducted an expert comparison. Two regions with distortions were shown to each participant: one area from the baseline algorithm and one from the machine learning model. The participants were asked to choose which region was better.

Comparison of the base algorithm and machine learning model


The most important features for scale mismatch

The most important features for scale mismatch

The most important features for rotation mismatch

The most important features for rotation mismatch

The most important features for sharpness mismatch

The most important features for sharpness mismatch

05 May 2020
See Also
Video Colorization Benchmark
Explore the best video colorization algorithms
MSU 3D-video Quality Analysis. Report 12
MSU 3D-video Quality Analysis. Report 11
MSU 3D-video Quality Analysis. Report 10
Detection of stereo window violation
How to find objects that are present only in one view?
Depth continuity estimation in S3D video
How smooth is the depth transition between scenes?
Site structure