One-Stage Object Detection Models
YOLOv4
Introduced by Bochkovskiy et al. in YOLOv4: Optimal Speed and Accuracy of Object DetectionYOLOv4 is a one-stage object detection model that improves on YOLOv3 with several bags of tricks and modules introduced in the literature. The components section below details the tricks and modules used.
Source: YOLOv4: Optimal Speed and Accuracy of Object Detection
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Object Detection | 44 | 32.12% |
| Real-Time Object Detection | 7 | 5.11% |
| Semantic Segmentation | 5 | 3.65% |
| Autonomous Driving | 4 | 2.92% |
| Domain Adaptation | 3 | 2.19% |
| Image Classification | 3 | 2.19% |
| Traffic Sign Detection | 3 | 2.19% |
| Optical Character Recognition (OCR) | 3 | 2.19% |
| Object Tracking | 3 | 2.19% |
Cosine Annealing
CSPDarknet53
CutMix
DropBlock
Label Smoothing
Mish
PAFPN
Spatial Attention Module
Spatial Pyramid Pooling
YOLOv3
