Click the Card below and follow the “CVer” public account
AI/CV heavy-duty information, delivered as soon as possible
Click to enter->[OpenCV] Communication Group
Reply in the background of CVer WeChat public account: OpenCV, you can download OpenCV learning materials and start learning!
Reprinted from: CSDN (ID: CSDNnews) | Tu Min
Without financial support, the development and maintenance of open source projects is difficult, and sometimes it is even impossible to see the future.
Recently, behind a poster “Keep Open Source Computer Vision & AI Alive” (let open source computer vision and AI “alive”), the veteran open source computer vision and machine learning software library OpenCV team issued Asking for help, with the theme of “The largest ever release of the world’s largest computer vision library, supporting open source”, we launched a crowdfunding campaign on the fundraising website Indiegogo, hoping to raise funds for the next version of OpenCV – OpenCV 5 required funds.
OpenCV has evolved for 23 years and has becomea widely used computer vision library in the AI field
“I hope to research some directions that are more valuable to the entire computer field. I found that computer vision requires a lot of calculations, and most laboratories are reinventing the wheel. Some of the work can even be done by graduate students in schools. Why not What about opening the library to everyone? This is the core reason why I created OpenCV,” OpenCV founder Gary Bradski said in an early interview with CSDN.
With this original intention, OpenCV was launched as an Intel research project in 1999, and then the first preview version was released in 2000, aiming to advance CPU-intensive applications, including a series of projects such as real-time ray tracing and 3D projection display.
As a product that follows the Apache 2 license, OpenCV allows users to easily use and modify code, which is one of the important reasons why OpenCV is widely used.
“From a data point of view, a search for OpenCV on Google Scholar returned 175,000 results, and a search on Google returned 43,700,000 results,” the OpenCV team wrote in the crowdfunding announcement. “According to industry reports, 89% of embedded OpenCV is used by modern vision engineers as the primary computer vision library, and the technology itself is used on probes and satellites in outer space, as well as in emergency rooms, farms, factories, and college campuses.”
Why does OpenCV 5 launch crowdfunding?
In fact, as an open source project, OpenCV cannot survive today without the support of some major technology companies.
OpenCV has also done some fundraising before, but it was mainly a crowdfunding campaign for some hardware and courses. Now, for the first time, a crowdfunding campaign has been launched for the development of the OpenCV library itself.
The reason for doing this is that OpenCV 5 is expected to be a breakthrough release, the OpenCV team said. The core team has been working hard to develop new advanced algorithms, expand to more platforms, enhance integration with other tools, and optimize for more different types of hardware. The proposed changes and additions are substantial and will require significant time and effort from the core team and Top Contributors on GitHub.
Unfortunately, due to changes in the global competitive environment, OpenCV has lost some core team members, which has slowed down its progress in developing new features and handling bugs and requests in its GitHub repository.
Under this, the OpenCV crowdfunding development of OpenCV 5 has also set three goals:
-
Making OpenCV 5 the largest release to date with the most community involvement;
-
Prove to other struggling open source projects that crowdfunding is a viable option that doesn’t rely solely on corporate donations;
-
Create a sustainable fundraising method that allows OpenCV to return each year, making the entire organization and project stronger in the face of global instability.
“Crowdfunding the development of a large open source project like OpenCV is a major initiative. We don’t know of anyone doing it on this scale. As a non-profit, we are constantly working to find funding, but we are always on our Find support in the community. Instead of spending time chasing big checks from billionaires, we put the community first,” said OpenCV.
As a result, OpenCV launched a crowdfunding campaign on the fundraising website Indiegogo on October 24. Phil Nelson, OpenCV’s content and creative director, said on the official blog:
“In this event, we will connect the entire community – including individual developers, small, medium and large companies, and creators. When we work together, we can do amazing things.”
How much money and resources will it cost to develop the next generation of OpenCV 5.0?
In fact, OpenCV 5.0 was originally planned as the 20th anniversary version and was initially planned to be released in 2020. However, if the latest OpenCV 5.0 wants to be implemented, as mentioned above, it faces huge challenges.
Currently, OpenCV has a crowdfunding goal of $500,000. At the same time, the team also listed on the crowdfunding platform the time and resource costs it will cost developers to develop each feature, such as:
-
Deep neural network module: $150,000
-
Accelerated image processing: $150,000
-
TrueType font support (including full Unicode): $50,000
-
Improved support for fisheye camera calibration: [$100,000 + hardware (~$20,000, shared with multi-camera support)]
-
Improved support for multi-camera calibration: [$100,000 + hardware (~$20,000, shared with fisheye cameras)]
-
Better support for depth cameras and new algorithms
-
Raspberry Pi Kit
-
Android Kit
-
Jetson Kit
-
Better optimization for AMD CPUs and GPUs
-
Add more models to the OpenCV model repository, such as Anomalib
-
Code refactoring/functional simplification: ($40,000)
-
Much improved documentation and updated tutorials
Stretch goals
In the vision of the OpenCV team, if this crowdfunding can exceed the goal of $500,000, they have something more they want to do.
-
When $600,000 is reached: Hire a full-time senior core team member.
-
When $700,000 is reached: Add a second continuous integration and build servers in a new location.
-
When $800,000 is reached: OpenCV AI Competition 2023 with a minimum prize of $100,000.
-
When $1M is reached: Freemium OpenCV cloud service for 3D computer vision. For example, Gaussian distribution training and inference.
Open source crowdfunding is also a difficult road
Finally, the OpenCV team guarantees that 100% of its crowdfunding funds will be used to hire engineers (part-time or full-time) to develop the OpenCV library. Depending on how much money is raised, it may also need to hire interns. Within OpenCV, the two main organizations that directly help with development are OpenCV.AI (headquartered in Cyprus) and OpenCV China, with contributors located around the world.
If the crowdfunding goes well, it will launch the OpenCV 5.0 version within 6-7 months after the crowdfunding. Based on the current time, it will probably be released in July or August 2024.
However, in reality, 10 days have passed since its launch of crowdfunding. As of now, OpenCV has only raised 4% of its US$500,000 goal, or US$22,740. There is still a long way to go to maintain the crowdfunding of open source projects. Far.
source:
https://www.indiegogo.com/projects/opencv-5-support-non-profit-open-source-cv-ai#/discussion
https://www.i-programmer.info/news/80-java/16662-opencv-launches-crowdfunding-campaign.html
Click to enter-> [OpenCV] Communication Group< /em>
ICCV/CVPR 2023 paper and code download
Backend reply: CVPR2023, you can download the CVPR 2023 papers and code open source paper collection
Backend reply: ICCV2023, you can download the collection of ICCV 2023 papers and code open source papers
OpenCV exchange group established Scan the QR code below, or add WeChat: CVer444 to add CVer assistant WeChat, and then apply to join the CVer-OpenCV WeChat exchange group. In addition, other vertical directions have been covered: target detection, image segmentation, target tracking, face detection & recognition, OCR, pose estimation, super-resolution, SLAM, medical imaging, Re-ID, GAN, NAS, depth estimation, automatic Driving, reinforcement learning, lane detection, model pruning & compression, denoising, fog removal, rain removal, style transfer, remote sensing images, behavior recognition, video understanding, image fusion, image retrieval, paper submission & communication , PyTorch, TensorFlow and Transformer, NeRF, etc. Be sure to note: research direction + location + school/company + nickname (such as OpenCV + Shanghai + hand in + Kaka). Note according to the format to be approved and invited to the group faster
▲Scan the QR code or add WeChat ID: CVer444 to join the communication group CVer Computer Vision (Knowledge Planet) is here! If you want to know about the latest, fastest and best CV/DL/AI paper express delivery, high-quality practical projects, AI industry cutting-edge, and learning tutorials from entry to mastery, please scan the QR code below and join CVer Computer Vision (Knowledge Planet). Nearly ten thousand people have been gathered! ▲Scan the QR code to join Planet Learning
▲Click on the card above to follow the CVer official account
It is not easy to organize, please like and watch