DDCNet: Deep Dilated Convolutional Neural Network for Dense Prediction
arXiv:2107.04715 · doi:10.1016/j.neucom.2022.12.024
Abstract
Dense pixel matching problems such as optical flow and disparity estimation are among the most challenging tasks in computer vision. Recently, several deep learning methods designed for these problems have been successful. A sufficiently larger effective receptive field (ERF) and a higher resolution of spatial features within a network are essential for providing higher-resolution dense estimates. In this work, we present a systemic approach to design network architectures that can provide a larger receptive field while maintaining a higher spatial feature resolution. To achieve a larger ERF, we utilized dilated convolutional layers. By aggressively increasing dilation rates in the deeper layers, we were able to achieve a sufficiently larger ERF with a significantly fewer number of trainable parameters. We used optical flow estimation problem as the primary benchmark to illustrate our network design strategy. The benchmark results (Sintel, KITTI, and Middlebury) indicate that our compact networks can achieve comparable performance in the class of lightweight networks.
arXiv comment32 pages, 13 figures, 3 tables
References in corpus (9)
- A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation
- Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
- Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
- Deep Predictive Coding Networks for Video Prediction and Unsupervised Learning
- Systematic evaluation of CNN advances on the ImageNet
- Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
- PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume
- Smoothed Dilated Convolutions for Improved Dense Prediction
- LiteFlowNet: A Lightweight Convolutional Neural Network for Optical Flow Estimation