4 ms·
Show HN: Computer Vision made easy with a new open-source Framework
- gdemarcq 4y agoFounder here, Ikomia API is an open source tool for building and deploying Computer Vision solutions without much effort. It's a Python library built upon a C++ core. You can mix your preferred frameworks such as OpenCV, Detectron2, OpenMMLab or YOLO with the best state-of-the-art algorithms from individual repos. At Ikomia, we deeply believe that sharing scientific knowledge is the key to success, that's why we try to make research-based algorithms ready-to-use for developers. Ikomia is developed by developers for developers :)
- FloatArtifact 4y agoIt would be interesting to see a benchmarking test suit for the algorithms supported in the context of their domain space. Not only good computer vision being made easy but also data-driven a synopsis allowing the developer to narrow in based on their applications needs.
- gdemarcq 4y agoSure! That's a very good remark and we plan to integrate this kind of information to our tool. For the moment, you can check paperswithcode.com which already gives some useful insights on benchmarking results ;)
- machiaweliczny 4y agoCan it be under Apache licence when YOLO is GPL?
- gdemarcq 4y agoEvery algorithms integrated in Ikomia HUB keep the License defined by its author. For example, YOLOv7 is GPL3 and our algorithm "infer_yolo_v7" is also GPL3. https://github.com/Ikomia-hub/infer_yolo_v7 https://github.com/Ikomia-hub/infer_yolo_v7 Algorithms in Ikomia HUB are not in Ikomia API code, they are used by the API. So if you use Ikomia API (Apache 2.0) with YOLOv7 (GPL3), your workflow becomes GPL3.