Train object detectors using images synthesized from real unmarked images
Train object detectors using images synthesized from real unmarked images
Self-Supervised Object Detection via Generative Image Synthesis
arXiv paper abstract https://arxiv.org/abs/2110.09848
arXiv PDF paper https://arxiv.org/pdf/2110.09848.pdf
... present SSOD, the first ... synthesis framework with controllable GANs for the task of self-supervised object detection.
... use collections of real world images without bounding box annotations to learn to synthesize and detect objects.
... leverage controllable GANs to synthesize images with pre-defined object properties and use them to train object detectors.
... also propose a method to optimally adapt SSOD to an intended target data without requiring labels for it.
For ... car detection ... show that SSOD outperforms the prior state-of-the-art purely image-based self-supervised object detection method Wetectron.
... without requiring any 3D CAD assets, it also surpasses the state-of-the-art rendering based method Meta-Sim2. ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments