A Unified Unsupervised Learning Framework for Stereo Matching and Ego-Motion Estimation

This paper presents a depth upsampling method that produces a high-fidelity dense depth map using a high-resolution RGB image and LiDAR sensor data. Our proposed method explicitly handles depth outliers and computes a depth upsampling with confidence information. Our key idea is the self-learning framework, which automatically learns to estimate the reliability of the upsampled depth map without human-labeled annotation. Thereby, our proposed method can produce a clear and high-fidelity dense depth map that preserves the shape of object structures well, which can be favored by subsequent algorithms for follow-up tasks. We qualitatively and quantitatively evaluate our proposed method by comparing other competing methods on the well-known Middlebury 2014 and KITTIbenchmark datasets. We demonstrate that our method generates accurate depth maps with smaller errors favorable against other methods while preserving a larger number of valid points, as we also show that our approach can be seamlessly applied to improve the quality of depth maps from other depth generation algorithms such as stereo matching and further discuss potential applications and limitations. Compared to previous work, our proposed method has similar depth errors on average, while retaining at least 3% more valid depth points.

Download Full-text

Review of Stereo Matching Algorithms Based on Deep Learning

Computational Intelligence and Neuroscience ◽

10.1155/2020/8562323 ◽

2020 ◽

Vol 2020 ◽

pp. 1-12 ◽

Cited By ~ 1

Author(s):

Kun Zhou ◽

Xiangxi Meng ◽

Bo Cheng

Keyword(s):

Deep Learning ◽

Unsupervised Learning ◽

Stereo Vision ◽

Stereo Matching ◽

Learning Algorithms ◽

Time Consumption ◽

End To End ◽

Traditional Approaches ◽

Speed Accuracy ◽

Comprehensive Coverage

Stereo vision is a flourishing field, attracting the attention of many researchers. Recently, leveraging on the development of deep learning, stereo matching algorithms have achieved remarkable performance far exceeding traditional approaches. This review presents an overview of different stereo matching algorithms based on deep learning. For convenience, we classified the algorithms into three categories: (1) non-end-to-end learning algorithms, (2) end-to-end learning algorithms, and (3) unsupervised learning algorithms. We have provided a comprehensive coverage of the remarkable approaches in each category and summarized the strengths, weaknesses, and major challenges, respectively. The speed, accuracy, and time consumption were adopted to compare the different algorithms.

Download Full-text