Corpus ID: 227276338

Fine-Grained Dynamic Head for Object Detection

@article{Song2020FineGrainedDH,
  title={Fine-Grained Dynamic Head for Object Detection},
  author={L. Song and Yanwei Li and Zhengkai Jiang and Zeming Li and Hongbin Sun and Jian Sun and Nanning Zheng},
  journal={ArXiv},
  year={2020},
  volume={abs/2012.03519}
}
The Feature Pyramid Network (FPN) presents a remarkable approach to alleviate the scale variance in object representation by performing instance-level assignments. Nevertheless, this strategy ignores the distinct characteristics of different sub-regions in an instance. To this end, we propose a fine-grained dynamic head to conditionally select a pixel-level combination of FPN features from different scales for each instance, which further releases the ability of multi-scale feature… Expand

Figures and Tables from this paper

IQDet: Instance-wise Quality Distribution Sampling for Object Detection
Rethinking Learnable Tree Filter for Generic Feature Transform
Dynamic Modality Interaction Modeling for Image-Text Retrieval
OTA: Optimal Transport Assignment for Object Detection
Large-scale image search with text for information retrieval
Dynamic Feature Pyramid Networks for Object Detection
End-to-End Object Detection with Fully Convolutional Network

References

SHOWING 1-10 OF 49 REFERENCES
Hypercolumns for object segmentation and fine-grained localization
Feature Pyramid Networks for Object Detection
Feature Pyramid Grids
Learning Dynamic Routing for Semantic Segmentation
  • Yanwei Li, L. Song, +4 authors J. Sun
  • Computer Science
  • 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2020
Path Aggregation Network for Instance Segmentation
Feature Selective Networks for Object Detection
NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection
Panoptic Feature Pyramid Networks
...
1
2
3
4
5
...