LeanStereo: A Leaner Backbone based Stereo Network
This work addresses the need for faster stereo matching in real-world applications, though it is incremental as it builds on existing deep learning approaches with optimizations.
The paper tackles the problem of high computational and memory requirements in deep stereo matching networks by proposing a leaner backbone combined with learned attention weights and LogL1 loss, resulting in a method that is 4x less operations and 9-14x faster than state-of-the-art while maintaining comparable performance.
Recently, end-to-end deep networks based stereo matching methods, mainly because of their performance, have gained popularity. However, this improvement in performance comes at the cost of increased computational and memory bandwidth requirements, thus necessitating specialized hardware (GPUs); even then, these methods have large inference times compared to classical methods. This limits their applicability in real-world applications. Although we desire high accuracy stereo methods albeit with reasonable inference time. To this end, we propose a fast end-to-end stereo matching method. Majority of this speedup comes from integrating a leaner backbone. To recover the performance lost because of a leaner backbone, we propose to use learned attention weights based cost volume combined with LogL1 loss for stereo matching. Using LogL1 loss not only improves the overall performance of the proposed network but also leads to faster convergence. We do a detailed empirical evaluation of different design choices and show that our method requires 4x less operations and is also about 9 to 14x faster compared to the state of the art methods like ACVNet [1], LEAStereo [2] and CFNet [3] while giving comparable performance.