READING NOTE: Beyond Skip Connections: Top-Down Modulation for Object Detection
2016-12-21 18:45
1146 查看
TITLE: Beyond Skip Connections: Top-Down Modulation for Object Detection
AUTHOR: Abhinav Shrivastava, Rahul Sukthankar, Jitendra Malik, Abhinav Gupta
ASSOCIATION: CMU, UC Berkeley, Google Research
FROM: arXiv:1612.06851
![](https://raw.githubusercontent.com/joshua19881228/my_blogs/master/Computer_Vision/Reading_Note/figures/TDM_1.jpg)
TDM is integrated with the bottom-up network with lateral connections. Ci are bottom-up, feedforward feature blocks, Li are the lateral modules which transform low level features for the top-down contextual pathway. Finally, Tj,i, which represent flow of top-down information from index j to i.
In this paper, the T blocks are implemented using single convolutional layer (with non-linear activation) optionally with upsampling operation. The features from C (processed by L) and T are concated then sent to a convolutional layer for combination, as the following figure shows
![](https://raw.githubusercontent.com/joshua19881228/my_blogs/master/Computer_Vision/Reading_Note/figures/TDM_2.jpg)
At training stage, one new pair of lateral and top-down modules is added at a time and trained repeatedly from a pre-trained model.
AUTHOR: Abhinav Shrivastava, Rahul Sukthankar, Jitendra Malik, Abhinav Gupta
ASSOCIATION: CMU, UC Berkeley, Google Research
FROM: arXiv:1612.06851
CONTRIBUTIONS
In this paper top-down modulations is proposed as a way to incorporate fine details into the detection framework. The standard bottom-up, feedforward ConvNet is supplemented with a top-down modulation (TDM) network, connected using lateral connections. These connections are responsible for the modulation of lower layer filters, and the top-down network handles the selection and integration of features.METHOD
The idea of this work is very similar with the work of Feature Pyramid Networks for Object Detection. An example of Top-Down Modulation (TDM) Network is illustrated as the following figure![](https://raw.githubusercontent.com/joshua19881228/my_blogs/master/Computer_Vision/Reading_Note/figures/TDM_1.jpg)
TDM is integrated with the bottom-up network with lateral connections. Ci are bottom-up, feedforward feature blocks, Li are the lateral modules which transform low level features for the top-down contextual pathway. Finally, Tj,i, which represent flow of top-down information from index j to i.
In this paper, the T blocks are implemented using single convolutional layer (with non-linear activation) optionally with upsampling operation. The features from C (processed by L) and T are concated then sent to a convolutional layer for combination, as the following figure shows
![](https://raw.githubusercontent.com/joshua19881228/my_blogs/master/Computer_Vision/Reading_Note/figures/TDM_2.jpg)
At training stage, one new pair of lateral and top-down modules is added at a time and trained repeatedly from a pre-trained model.
相关文章推荐
- 目标检测--Beyond Skip Connections: Top-Down Modulation for Object Detection
- 解读Top-Down Modulation for object detection
- READING NOTE: PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
- Reading Note: Single-Shot Refinement Neural Network for Object Detection
- READING NOTE: Feature Pyramid Networks for Object Detection
- READING NOTE: Object Detection by Labeling Superpixels
- Beyond Skip Connections: Top-Down Modulation for Object Detection阅读笔记
- Ensemble of Exemplar-SVMs for Object Detection and Beyond(2011)学习笔记
- [Paper note] Feature Pyramid Networks for Object Detection
- Exemplar-SVMs for Object Detection and Beyond—Overview(一)
- [Paper note] PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
- Exemplar-SVMs for Object Detection and Beyond--Details (二)
- READING NOTE: Object Detection from Video Tubelets with Convolutional Neural Networks
- Exemplar-SVMs for Object Detection and Beyond--Overview(一)
- READING NOTE: Pushing the Limits of Deep CNNs for Pedestrian Detection
- Exemplar-SVMs for Object Detection and Beyond—Details(二)
- READING NOTE: Spatially Supervised Recurrent Convolutional Neural Networks for Visual Object Trackin
- Exemplar-SVMs for Object Detection and Beyond—Overview(一)
- Exemplar-SVMs for Object Detection and Beyond--Details (二)
- READING NOTE: Speed/accuracy trade-offs for modern convolutional object detectors