Better small object detection by using Wasserstein distance with NWD
Better small object detection by using Wasserstein distance with NWD
A Normalized Gaussian Wasserstein Distance for Tiny Object Detection
arXiv paper abstract https://arxiv.org/abs/2110.13389
arXiv PDF paper https://arxiv.org/pdf/2110.13389.pdf
Detecting tiny objects is a very challenging problem since a tiny object only contains a few pixels in size.
... propose a new evaluation metric using Wasserstein distance for tiny object detection.
... first model the bounding boxes as 2D Gaussian distributions and then propose a new metric dubbed Normalized Wasserstein Distance (NWD) to compute the similarity between them by their corresponding Gaussian distributions.
... NWD metric can be easily embedded into the assignment, non-maximum suppression, and loss function of any anchor-based detector to replace the commonly used IoU metric.
... evaluate ... metric on a new dataset for tiny object detection (AI-TOD) in which the average object size is much smaller than existing object detection datasets.
... with NWD metric, ... performance that is 6.7 AP points higher than a standard fine-tuning baseline, and 6.0 AP points higher than state-of-the-art competitors.
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
Commenti