Train object detector by labeling only 1 point of object with NSS
Train object detector by labeling only 1 point of object with NSS
Weakly-Supervised Salient Object Detection Using Point Supervison
arXiv paper abstract https://arxiv.org/abs/2203.11652v1
arXiv PDf paper https://arxiv.org/pdf/2203.11652v1.pdf
Current ... saliency detection models rely ... on large datasets of accurate pixel-wise annotations, but manually labeling pixels is time-consuming
... some weakly supervised methods ... alleviating the problem, such as image label, bounding box label, and scribble label
... propose a novel weakly-supervised salient object detection method using point supervision. ... first design an adaptive masked flood filling algorithm to generate pseudo labels.
... develop a transformer-based point-supervised saliency detection model to produce the first round of saliency maps.
... propose a Non-Salient Suppression (NSS) method to optimize the erroneous saliency maps generated in the first round and leverage them for the second round of training.
... method outperforms ... state-of-the-art methods trained with the stronger supervision and even surpass several fully supervised state-of-the-art models. ...
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
Kommentare