Optimized (very fast) stereo matching algorithms in MATLAB and Python. It includes implementations of Block Matching, Dynamic Programming, Semi-Global Matching, Semi-Global Block Matching, Belief Propagation and Iterated Conditional Modes.
-
Stereo matching algorithms
- Block Matching
- Dynamic Programming
- Semi-Global Matching
- Semi-Global Block Matching
- Belief Propagation
- Iterated Conditional Modes
-
Multiple different versions of the algorithms
-
All algorithms are implemented in both MATLAB and Python
-
The algorithms are optimized for performance using matrix operations and other techniques
| Name | Description | MATLAB | Python |
|---|---|---|---|
| Block Matching (SAD) | Block Matching using Sum of Absolute Differences | stereoBM_SAD.m |
stereoBM_SAD.py |
| Block Matching (SAD) | Block Matching using Sum of Absolute Differences (different approach) | stereoBM_SAD2.m |
stereoBM_SAD2.py |
| Block Matching (Gradient) | Block Matching using Image Gradients | stereoBM_Grad.m |
stereoBM_Grad.py |
| Block Matching (NCC) | Block Matching using Normalized Cross-Correlation | stereoBM_NCC.m |
stereoBM_NCC.py |
| Block Matching (Rank) | Block Matching using Rank Transformation | stereoBM_Rank.m |
stereoBM_Rank.py |
| Block Matching (Census) | Block Matching using Census Transformation and Hamming Distance | stereoBM_Census.m |
stereoBM_Census.py |
| Block Matching (Adaptive) | Block Matching using Adaptive Window (Adaptive Support Weights) | stereoBM_ASW.m |
stereoBM_ASW.py |
| Dynamic Programming (Left-Right) | Dynamic Programming with Left–Right Axes DSI | stereoDP_LR.m |
stereoDP_LR.py |
| Dynamic Programming (Left-Disparity) | Dynamic Programming with Left–Disparity Axes DSI | stereoDP_LD.m |
stereoDP_LD.py |
| Semi-Global Matching | Semi-Global Matching with 8-path cost aggregation | stereoSGM.m |
stereoSGM.py |
| Semi-Global Block Matching | Semi-Global Block Matching with 8-path cost aggregation | stereoSGBM.m |
stereoSGBM.py |
| Belief Propagation (Accelerated) | Belief Propagation with Accelerated (or Directional) Message Update Schedule | stereoBP_Accel.m |
stereoBP_Accel.py |
| Belief Propagation (Synchronous) | Belief Propagation with Synchronous Message Update Schedule | stereoBP_Synch.m |
stereoBP_Synch.py |
| Belief Propagation (Synchronous) | Belief Propagation with Synchronous Message Update Schedule (different approach) | stereoBP_Synch2.m |
stereoBP_Synch2.py |
| Belief Propagation (Bipartite) | Belief Propagation with Bipartite Message Update Schedule | stereoBP_Bipart.m |
stereoBP_Bipart.py |
| Iterated Conditional Modes | Iterated Conditional Modes with Synchronous Update Schedule | stereoICM.m |
stereoICM.py |
Download the project as ZIP file, unzip it, and run the scripts.
- NumPy
- Matplotlib
- OpenCV (
opencv-python)
A stereo matching algorithm works with stereo image pairs to produce disparity maps.
This project contains MATLAB and Python scripts, each implementing a stereo matching algorithm. The files left.png and right.png contain the stereo image pair used as input.
To use a different stereo pair, replace these two images with your own. In this case, you must also adjust the disparity levels parameter in the script you are running.
You may optionally modify other parameters as needed. If the input images contain little or no noise, it is recommended not to use the Gaussian filter.
- The results between MATLAB and Python implementation are similar.
- The different approaches produce same results.
The following running times are in seconds and were measured on a Windows PC with a CPU AMD A10-7850K and 8GB of RAM.
| Filename | Running Time MATLAB | Running Time Python | Notes |
|---|---|---|---|
stereoBM_SAD |
0.10 | 0.08 | image display disabled |
stereoBM_SAD2 |
0.48 | 0.81 | image display disabled |
stereoBM_Grad |
0.17 | 0.19 | image display disabled |
stereoBM_NCC |
1.52 | 1.84 | image display disabled |
stereoBM_Rank |
0.77 | 3.70 | image display disabled |
stereoBM_Census |
2.94 | 7.25 | image display disabled |
stereoBM_ASW |
20.58 | 46.11 | image display disabled |
stereoDP_LR |
0.32 | 5.60 | image display disabled |
stereoDP_LD |
0.14 | 0.28 | image display disabled |
stereoSGM |
0.81 | 2.45 | image display disabled |
stereoSGBM |
0.84 | 2.51 | image display disabled |
stereoBP_Accel |
9.11 | 15.18 | image display disabled, 20 iterations |
stereoBP_Synch |
11.16 | 21.00 | image display disabled, 20 iterations |
stereoBP_Synch2 |
4.83 | 17.57 | image display disabled, 20 iterations |
stereoBP_Bipart |
23.38 | 45.79 | image display disabled, 20 iterations |
stereoICM |
1.95 | 5.90 | image display disabled, 20 iterations |
Below are the disparity maps produced from the Tsukuba stereo pair.
- Basic Stereo Algorithms (Evolution)
- Belief Propagation for Stereo Matching (Sum-Product, Max-Product, Min-Sum)
This project is licensed under the MIT License. See the LICENSE file for details.