Detect more objects when only using image-level labels with WS-DETR
Detect more objects when only using image-level labels with WS-DETR
Scaling Novel Object Detection with Weakly Supervised Detection Transformers
arXiv paper abstract https://arxiv.org/abs/2207.05205
arXiv PDF paper https://arxiv.org/pdf/2207.05205.pdf
Weakly supervised object detection (WSOD) enables object detectors to be trained using image-level class labels.
However, the practical application of current WSOD models is limited, as they operate at small scales and require extensive training and refinement.
... propose the Weakly Supervised Detection Transformer, which enables efficient knowledge transfer from a large-scale pretraining dataset to WSOD finetuning on hundreds of novel objects.
... leverage pretrained knowledge to improve the multiple instance learning framework used in WSOD, and experiments show
... approach outperforms the state-of-the-art on datasets with twice the novel classes than previously shown.
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