Local Similarity Pattern and Cost Self-Reassembling for Deep Stereo Matching Networks
This work addresses over-smoothing and feature discrimination issues in stereo matching for computer vision applications, representing an incremental advancement by integrating traditional wisdoms into existing architectures.
The paper tackled the limitations of convolutional features and static filters in stereo matching networks by introducing Local Similarity Pattern (LSP) for richer structural information and a dynamic self-reassembling refinement strategy for cost distribution and disparity maps, resulting in significant performance improvements on SceneFlow and KITTI benchmarks.
Although convolution neural network based stereo matching architectures have made impressive achievements, there are still some limitations: 1) Convolutional Feature (CF) tends to capture appearance information, which is inadequate for accurate matching. 2) Due to the static filters, current convolution based disparity refinement modules often produce over-smooth results. In this paper, we present two schemes to address these issues, where some traditional wisdoms are integrated. Firstly, we introduce a pairwise feature for deep stereo matching networks, named LSP (Local Similarity Pattern). Through explicitly revealing the neighbor relationships, LSP contains rich structural information, which can be leveraged to aid CF for more discriminative feature description. Secondly, we design a dynamic self-reassembling refinement strategy and apply it to the cost distribution and the disparity map respectively. The former could be equipped with the unimodal distribution constraint to alleviate the over-smoothing problem, and the latter is more practical. The effectiveness of the proposed methods is demonstrated via incorporating them into two well-known basic architectures, GwcNet and GANet-deep. Experimental results on the SceneFlow and KITTI benchmarks show that our modules significantly improve the performance of the model.