티스토리 뷰
[2018][***]Small Object Detection Using Deep Feature Pyramid Networks
Arc Lab. 2019. 7. 31. 11:151. 인용 논문
Traffic-sign detection and classification in the wild
Zhe Zhu, Dun Liang, Songhai Zhang, Xiaolei Huang, Baoli Li, Shimin Hu;
The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 2110-2118
2. 인용 부분
- Page 4-5
To detect small objects from a high-resolution image, we design a two-stage architecture as shown in Fig. 3. For region proposal stage, two top-down feature pyramids are built to obtain the high-level semantic feature maps at all scales, which boost the discriminative power of feature representation for small object.
For classification stage, we adopt the idea of Densenet [6]. To improve accuracy, focal loss [14] is used to supervise the proposal network.
Focal loss [14] achieves great success on solving the problem of class imbalance during training.

- Page 7

- Page 8

- Total
- Today
- Yesterday
- #REST API
- SSM
- English
- #ELK Stack
- belief
- Game Engine
- Meow
- GOD
- ate
- 도커
- Memorize
- Sea Bottom
- 2D Game
- ILoop Engine
- project
- Library
- OST
- Mask R-CNN
- Jekyll and Hyde
- Badge
- docker
- #TensorFlow
- some time ago
- #ApacheZeppelin
- Worry
- #ApacheSpark
- Physical Simulation
- sentence test
- Ragdoll
- aws #cloudfront
일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | ||
6 | 7 | 8 | 9 | 10 | 11 | 12 |
13 | 14 | 15 | 16 | 17 | 18 | 19 |
20 | 21 | 22 | 23 | 24 | 25 | 26 |
27 | 28 | 29 | 30 | 31 |