Abstract
In recent years, disparity estimation of a scene based on deep learning methods has been extensively studied and significant progress has been made. In contrast, a traditional image disparity estimation method requires considerable resources and consumes much time in processes such as stereo matching and 3D reconstruction. At present, most deep learning based disparity estimation methods focus on estimating disparity based on monocular images. Motivated by the results of traditional methods that multi-view methods are more accurate than monocular methods, especially for scenes that are textureless and have thin structures, in this paper, we present MDEAN, a new deep convolutional neural network to estimate disparity using multi-view images with an asymmetric encoder–decoder network structure. First, our method takes an arbitrary number of multi-view images as input. Next, we use these images to produce a set of plane-sweep cost volumes, which are combined to compute a high quality disparity map using an end-to-end asymmetric network. The results show that our method performs better than state-of-the-art methods, in particular, for outdoor scenes with the sky, flat surfaces and buildings.
Highlights
Disparity estimation from images is playing an increasingly important role in computer vision.Numerous important applications, including 3D reconstruction, autonomous driving, robotics and medical image processing, require depth information
We present a new convolutional neural networks (CNNs)-based method called MDEAN which stands for a multi-view disparity estimation method using an asymmetric network
We propose an asymmetric fully convolutional network based on the encoder–decoder structure for scene disparity estimation
Summary
Disparity estimation from images is playing an increasingly important role in computer vision. Thanks to multi-view based methods, depth can be computed with high precision. Conventional multi-view depth estimation methods, such as Structure-from-Motion (SFM). Besides SFM, there are many conventional Multi-View Stereo (MVS) methods to estimate the depth map by computing the cost volume using the plane-sweep method [3] or by measuring the similarity between patches using some error functions [4]. We present a new CNN-based method called MDEAN which stands for a multi-view disparity estimation method using an asymmetric network. Based on the experimental results, our proposed method can obtain high-quality disparity maps (See Figure 1) It outperforms the state-of-the-art methods with better prediction results, in particular, in outdoor scenes with the sky, flat surfaces, buildings
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
