Corpus ID: 228083601

OneNet: Towards End-to-End One-Stage Object Detection

@article{Sun2020OneNetTE,
  title={OneNet: Towards End-to-End One-Stage Object Detection},
  author={Peize Sun and Yi Jiang and Enze Xie and Ze-Huan Yuan and C. Wang and Ping Luo},
  journal={ArXiv},
  year={2020},
  volume={abs/2012.05780}
}
End-to-end one-stage object detection trailed thus far. This paper discovers that the lack of classification cost between sample and ground-truth in label assignment is the main obstacle for one-stage detectors to remove Non-maximum Suppression(NMS) and reach end-to-end. Existing one-stage object detectors assign labels by only location cost, e.g. box IoU or point distance. Without classification cost, sole location cost leads to redundant boxes of high confidence scores in inference, making… Expand
6 Citations

Figures and Tables from this paper

QueryInst: Parallelly Supervised Mask Query for Instance Segmentation
  • Yuxin Fang, Shusheng Yang, +5 authors Wenyu Liu
  • Computer Science
  • 2021
  • PDF
RMOPP: Robust Multi-Objective Post-Processing for Effective Object Detection
  • PDF
Object Detection Made Simpler by Eliminating Heuristic NMS
  • PDF
Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
  • 21
  • PDF
SparsePoint: Fully End-to-End Sparse 3D Object Detector
  • Zili Liu, Guodong Xu, Honghui Yang, Haifeng Liu, Deng Cai
  • Computer Science
  • ArXiv
  • 2021
  • PDF
TransTrack: Multiple-Object Tracking with Transformer
  • 17
  • PDF

References

SHOWING 1-10 OF 46 REFERENCES
Corner Proposal Network for Anchor-free, Two-stage Object Detection
  • 11
  • PDF
You Only Look Once: Unified, Real-Time Object Detection
  • 11,531
  • Highly Influential
  • PDF
CenterNet: Keypoint Triplets for Object Detection
  • 386
  • PDF
Objects as Points
  • 497
  • Highly Influential
  • PDF
FCOS: Fully Convolutional One-Stage Object Detection
  • 640
  • Highly Influential
  • PDF
Focal Loss for Dense Object Detection
  • 4,439
  • Highly Influential
Focal Loss for Dense Object Detection
  • 2,544
  • PDF
DenseBox: Unifying Landmark Localization with End to End Object Detection
  • 238
  • PDF
Cascade R-CNN: Delving Into High Quality Object Detection
  • 1,032
  • PDF
Associative Embedding: End-to-End Learning for Joint Detection and Grouping
  • 385
  • PDF
...
1
2
3
4
5
...