Dynamic Filtering with Large Sampling Field for ConvNets
arXiv:1803.07624
Abstract
We propose a dynamic filtering strategy with large sampling field for ConvNets (LS-DFN), where the position-specific kernels learn from not only the identical position but also multiple sampled neighbor regions. During sampling, residual learning is introduced to ease training and an attention mechanism is applied to fuse features from different samples. Such multiple samples enlarge the kernels' receptive fields significantly without requiring more parameters. While LS-DFN inherits the advantages of DFN, namely avoiding feature map blurring by position-wise kernels while keeping translation invariance, it also efficiently alleviates the overfitting issue caused by much more parameters than normal CNNs. Our model is efficient and can be trained end-to-end via standard back-propagation. We demonstrate the merits of our LS-DFN on both sparse and dense prediction tasks involving object detection, semantic segmentation, and flow estimation. Our results show LS-DFN enjoys stronger recognition abilities in object detection and semantic segmentation tasks on VOC benchmark and sharper responses in flow estimation on FlyingChairs dataset compared to strong baselines.
arXiv commentECCV 2018
References in corpus (19)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- R-FCN: Object Detection via Region-based Fully Convolutional Networks
- Unsupervised Learning of Video Representations using LSTMs
- Unsupervised Domain Adaptation with Residual Transfer Networks
- Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
- Multiple Object Recognition with Visual Attention
- Feature Pyramid Networks for Object Detection
- Deformable Convolutional Networks
- Action Recognition using Visual Attention
- Residual Attention Network for Image Classification
- Multimodal Residual Learning for Visual QA
- Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks
- FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks
- Fully Convolutional Instance-aware Semantic Segmentation
- Optical Flow Estimation using a Spatial Pyramid Network
- Action Recognition with Joint Attention on Multi-Level Deep Features