1. SOLOv2: Dynamic and Fast Instance Segmentation;wang;Advances in neural information processing systems,2020
2. Masked autoencoders are scalable vision learners;he;ArXiv,2021
3. Focal Loss for Dense Object Detection
4. Scaling Vision Transformers;zhai;ArXiv,2021
5. Microsoft COCO: Common Objects in Context;lin;European Conference on Computer Vision,2014