Focal Loss for Dense Object Detection

Tsung-Yi Lin1, Priya Goyal1, Ross Girshick1, Kaiming He1, Piotr Dollár1
1Facebook AI Research#TAB#

Tóm tắt

Từ khóa


Tài liệu tham khảo

girshick, 2008, Detectron

szegedy, 2017, Inception-v4, inception-resnet and the impact of residual connections on learning, Proc AAAI Conf Artif Intell

10.1109/CVPR.2017.351

sermanet, 2014, Overfeat: Integrated recognition, localization and detection using convolutional networks, Proc Int Conf Learn Representations

10.1109/CVPR.2016.90

shrivastava, 2016, Beyond skip connections: Top-down modulation for object detection, arXiv 1612 06851

10.1109/CVPR.2016.314

10.1109/CVPR.2015.7298965

hastie, 2008, The Elements of Statistical Learning

10.1109/CVPR.2017.749

fu, 2016, DSSD: Deconvolutional single shot detector, arXiv 1701 06659

10.1109/CVPR.2017.634

dai, 2016, R-FCN: Object detection via region-based fully convolutional networks, Proc Neural Inf Process Syst

10.1007/s11263-013-0620-5

zitnick, 2014, Edge boxes: Locating object proposals from edges, Proc Eur Conf Comput Vis, 391

pinheiro, 2015, Learning to segment object candidates, Proc Neural Inf Process Syst

pinheiro, 2016, Learning to refine object segments, Proc Eur Conf Comput Vis

10.1109/CVPR.2016.89

sung, 1994, Learning and example selection for object and pattern detection

rowley, 1995, Human face detection in visual scenes

10.1109/CVPR.2001.990517

10.1109/TPAMI.2015.2465908

10.1109/CVPR.2017.106

krizhevsky, 2012, ImageNet classification with deep convolutional neural networks, Proc Neural Inf Process Syst, 1097

ren, 2015, Faster R-CNN: Towards real-time object detection with region proposal networks, Proc Neural Inf Process Syst

lin, 2014, Microsoft COCO: Common objects in context, Proc Eur Conf Comput Vis, 740

he, 2014, Spatial pyramid pooling in deep convolutional networks for visual recognition, Proc Eur Conf Comput Vis, 346

10.1109/ICCV.2017.322

10.1109/CVPR.2017.690

10.1109/CVPR.2016.91

10.1109/ICCV.2015.169

liu, 2016, SSD: Single shot multibox detector, Proc Eur Conf Comput Vis

10.1109/CVPR.2014.81

10.1109/CVPR.2010.5539906

10.1162/neco.1989.1.4.541

10.1109/CVPR.2014.276

10.1109/CVPR.2005.177

10.1049/ip-vis:19941301

10.1007/s11263-009-0275-4

10.5244/C.23.91