Git Product home page Git Product logo

prime-project / primestereomatch Goto Github PK

View Code? Open in Web Editor NEW
222.0 222.0 65.0 28.19 MB

A heterogeneous and fully parallel stereo matching algorithm for depth estimation, implementing a local adaptive support weight (ADSW) Guided Image Filter (GIF) cost aggregation stage. Developed in both C++ and OpenCL.

License: BSD 3-Clause "New" or "Revised" License

C 10.72% C++ 88.77% CMake 0.51%
algorithm c c-plus-plus camera computer-vision depth-estimation disparity-map opencl opencv stereo-matching

primestereomatch's People

Contributors

cleechelec avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

primestereomatch's Issues

HCI: Error threshold set to 4.

I am run ./PRiMEStereoMatch image -l ./left.png -r ./right.png --alg=STEREO_SGBM
image

OpenCL Platform Information:

  1. Device: GeForce GTX 960M
    1.0 Device Type: GPU
    1.1 Hardware version: OpenCL 1.2 CUDA
    1.2 Software version: 384.130
    1.3 OpenCL C version: OpenCL C 1.2
    1.4 Parallel compute units: 5
    1.5 Max Work Group Size: 0
    1.6 Max Work Item Dimensions: 3
    1.7 Max Global Memory Size: 4240965632
    1.8 Image Support?: Yes

Starting Stereo Matching Application.
Stereo Matching for Depth Estimation.
Preprocessing for Stereo Matching.
Input Source: Images
Parsing command-specific arguments.
User Dataset filenames provided.
Image command arguments parsed.
Global arguments:
Matching Algorithm: STEREO_SGBM
Loading User as the default dataset.
Loaded Left: ./left.png
Loaded Right: ./right.png
Setting up pthreads function constructors
Setting up OpenCL Environment
Number of devices available of each type:
CL_DEVICE_TYPE_CPU: 0
CL_DEVICE_TYPE_GPU: 1
CL_DEVICE_TYPE_ACCELERATOR: 0
-----CL_DEVICE_TYPE_GPU---------
Setting up OpenCL function constructors
Build log:

CVC_cl: OpenCL kernels created.
Build log:

CVF_cl: OpenCL kernel versions created in context.
Allocated OpenCL Buffers
Build log:

DispSel_cl: OpenCL kernels created.
Construction Complete
End of Preprocessing
StereoMatch Application Initialised
Computing Depth Map
MatchingAlgorithm == STEREO_SGBM
HCI: Error threshold set to 4.
DE Time: 155.65 ms

CL_MEM_OBJECT_ALLOCATION_FAILURE when clEnqueueNDRangeKernel

Run time: 3.61152ms
CVF_cl: Running central filter Kernels
OpenCL error: CL_MEM_OBJECT_ALLOCATION_FAILURE
Failed enqueuing the kernel. src/CVF_cl.cpp:651
CVF_cl: Running boxfilter Kernels
Segmentation fault (core dumped)

Platform Name = NVIDIA CUDA, Platform Version = OpenCL 1.2 CUDA 7.5.26, SDK Revision = 7027912, NumDevs = 1, Device = GeForce GTX 645

System Info:
CPU Name: Intel(R) Core(TM) i5-4460 CPU @ 3.20GHz

of CPU processors: 4

Linux version 3.19.0-43-generic (buildd@lgw01-16) (gcc version 4.8.2 (Ubuntu 4.8.2-19ubuntu1) ) #49~14.04.1-Ubuntu SMP Thu Dec 31 15:44:49 UTC 2015

STEREO_GIF with OpenCL

Hello,
I've been working on trying to get the STEREO_GIF algorithm to run with OpenCL. It doesn't seem to work directly out of the box. Has anyone had success accomplishing this? It seems like the non-OpenCL stuff runs great.
Z

Need instructions for using Opencl based stereo matching on android

I already have calibrated and rectified left and right image as opencv Mat. I am currently using sgbm module for finding depth map and for post processing.

What is expected fps for a disparity map for image 640x480 with post processing?

Can you provide instructions for which functions of your code (minimalist) if I need opencl based stereo matching and post processing, which I plan to run on Odroid SBC?

Output window freezes!

On Ubuntu 16.04, after building, and using this command below the output window appears and freezes...

./PRiMEStereoMatch image -l "/media/mohammad/새 볼륨/STEREO/PRiMEStereoMatch/data/Art/view1.png" -r "/media/mohammad/새 볼륨/STEREO/PRiMEStereoMatch/data/Art/view5.png" -a STEREO_SGBM
./PRiMEStereoMatch: /usr/local/cuda/lib64/libOpenCL.so.1: no version information available (required by ./PRiMEStereoMatch)

OpenCL Platform Information:

  1. Device: GeForce GTX 1060 3GB
    1.0 Device Type: GPU
    1.1 Hardware version: OpenCL 1.2 CUDA
    1.2 Software version: 384.111
    1.3 OpenCL C version: OpenCL C 1.2
    1.4 Parallel compute units: 9
    1.5 Max Work Group Size: 0
    1.6 Max Work Item Dimensions: 3
    1.7 Max Global Memory Size: 3158900736
    1.8 Image Support?: Yes

Starting Stereo Matching Application.
Input Source: Images
Parsing command-specific arguments.
User Dataset filenames provided.
Image command arguments parsed.
Global arguments:
Matching Algorithm: STEREO_SGBM
Loading User as the default dataset.
Loaded Left: /media/mohammad/새 볼륨/STEREO/PRiMEStereoMatch/data/Art/view1.png
Loaded Right: /media/mohammad/새 볼륨/STEREO/PRiMEStereoMatch/data/Art/view5.png
Setting up pthreads function constructors
Setting up OpenCL Environment
Number of devices available of each type:
CL_DEVICE_TYPE_CPU: 0
CL_DEVICE_TYPE_GPU: 1
CL_DEVICE_TYPE_ACCELERATOR: 0
Setting up OpenCL function constructors
Build log:

CVC_cl: OpenCL kernels created.
Build log:

CVF_cl: OpenCL kernel versions created in context.
Allocated OpenCL Buffers
Build log:

DispSel_cl: OpenCL kernels created.
Construction Complete
End of Preprocessing
StereoMatch Application Initialised
HCI: Error threshold set to 4.
DE Time: 49.15 ms
DE Time: 49.15 ms

image

BTW runninag as sudo since it gives OpenCL error: CL_OUT_OF_RESOURCES error if not.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.