mirror/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2024-11-24 03:00:14 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Updated GSoC_2019 (markdown)

Gary Bradski 2019-03-24 23:10:49 -07:00
parent 2af2cadfca
commit 1ab34f9dcd
1 changed files with 1 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
GSoC_2019.md

@ -17,7 +17,7 @@ OpenCV Project Ideas List:
| [Model Zoo](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-model-zoo-curate-deep-nets-for-the-DNN) | [Point Coordinate Regression](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-point-coordinate-regression-and-dsac) | [Differential Rendering](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-create-a-differentiable-rendering-module-torchgeometryrender) | [Image Processing](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-differentiable-image-processing) |
|[April Tags](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-add-april-tag-fiducial-detection-to-opencv) or [Geometric April Tags](https://github.com/opencv/opencv/wiki/GSoC_2019#idea-3d-augmentation--april-tags)| [Optical Flow](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-create-a-consolidated-optical-flow-module-optflow)| [Python OpenCV](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-improve-and-expand-python-version-of-opencv)| [Depth Fusion](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-kinectfusion-improvements)|
|[Face Landmarks](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-facial-landmark-detector)| [Boosted Cascades](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-a-universal-boosted-cascades-interface)| [Machine Learning Tutorials](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-improve-machine-learning-tutorials-and-documentation)| [OpenCV GUI](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-improve-the-opencv-user-interface)|
| [DNN Quantization](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-add-quantization-and-pruning-functionality-opencv-dnn-module) | [DNN Super-Resolution](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-deep-learning-based-super-resolution-algorithms-based-on-opencv-dnn) | [Alpha Matting](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-Alpha-Matting-algorithms) [**1**]() [**2**]()| [Deep Nets on Video](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-python-deep-learning-inference-on-video) |
| [DNN Quantization](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-add-quantization-and-pruning-functionality-opencv-dnn-module) | [DNN Super-Resolution](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-deep-learning-based-super-resolution-algorithms-based-on-opencv-dnn) | [Alpha Matting](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-deep-learning-based-alpha-matting) [**1**](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-deep-learning-based-alpha-matting) [**2**](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-computer-vision-based-Alpha-Matting)| [Deep Nets on Video](https://github.com/opencv/opencv/wiki/GSoC_2019/#idea-python-deep-learning-inference-on-video) |
| geometric vision | | |
-----------------
@ -200,21 +200,6 @@ OpenCV is taking part in [GSoC 2019](https://summerofcode.withgoogle.com/organiz
* ***Mentors:*** Tyan Vladimir
* ***Difficulty:*** Medium
1. #### _IDEA:_ Deep learning based Alpha Matting
* ***Description:*** Project aims at generating a large training dataset for alpha matting, training the model and converting the model to ONNX to be used with OpenCV's DNN module.
* ***Expected Outcomes:***
* A large dataset which can be used to train models of alpha matting.
* Implement the paper: ["Deep Image Matting"](https://arxiv.org/abs/1703.03872) by Ning Xu et al.
* Trained model by the use of the generated dataset
* Experiments comparing results to existing alpha matting algorithms at [alphamatting.com](http://alphamatting.com/index.html)
* Convert model to ONNX and provide running example in OpenCV's DNN module
* Resources
* [alphamatting.com](http://alphamatting.com/index.html) Comparison of many methods, datasets etc
* ***Skills Required:*** Excellent C++ and Python coding skills, Experience in training deep learning models in Tensorflow/Pytorch, Experience in Computer Vision, Experience in Deep Learning
* ***Mentors:*** Sunita Nayak
* ***Difficulty:*** Medium-Hard
1. #### _IDEA:_ Computer Vision based Alpha Matting
* ***Description:*** Project aims to integrate some of the best computer vision based best alpha matting algorithms into OpenCV.
* ***Expected Outcomes:***