基于多尺度聚合神经网络的双目视觉立体匹配方法

Binocular stereo matching method based on multi-scale aggregation neural network

为了改善机器人、无人驾驶领域采用深度神经网络实现双目视觉立体匹配存在参数量大、GPU资源成本高的问题,提出一种多尺度聚合的立体匹配方法。首先设计了一个结合多尺度的特征提取网络,利用空洞卷积在不改变分辨率下获得更为丰富的特征,引入注意力机制,再将不同分辨率下特征交叉融合以完善特征信息;其次,改变代价卷获取方式,在低尺度下聚合得到代价卷,不断结合高尺度相似信息以迭代更新,将多个代价卷进行交叉融合以得到最终代价卷;最后,结合注意力机制的精细化模块修正初始视差图中的异常值与不连续区域,得到最终视差图。实验结果表明,该算法能够在较低参数量,以及低成本GPU资源下运行,且获得较好的匹配精度。

In order to solve the problems of large number of parameters and high GPU resource cost in binocular vision stereo matching method based on neural network in robot and unmanned driving fields,this paper proposed a multi-scale aggregation stereo matching method.Firstly,this paper proposed a multi-scale feature extraction network to obtain richer features without changing the resolution by using dilated convolution,and introduced the attention mechanism.Then,it cross-fused features at different resolutions to improve the feature information.Secondly,the acquisition method of cost volume was changed,the cost volume was obtained by aggregation at low scale,and continuously combined the high-scale similar information to update iteratively,and cross fused multiple cost volumes to obtain the final cost volume.Finally,combined with the refinement module of attention mechanism,the outliers and discontinuous regions in the initial disparity map were corrected to obtain the final dispa-rity map.Experimental results show that the algorithm can run under low parameter number and low cost GPU resources,and obtain good matching accuracy.