DeepVideoMVS: Multi-View Stereo on Video with Recurrent Spatio-Temporal Fusion

Düzçeker, Arda; Galliani, Silvano; Vogel, Christoph S.; Speciale, Pablo; Dusmanu, Mihai; Pollefeys, Marc

doi:10.48550/arxiv.2012.02177

Cited by 1 publication

(2 citation statements)

References 39 publications

(72 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Approaches for performing this task are varied [ 36 , 37 ] and are, by their nature, often unsuitable for real-time depth estimation. However, some are adaptable for depth estimation on sequences [ 38 , 39 , 40 ], while others are specifically designed to work on image sequences in real time [ 41 , 42 ].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Parallax Inference for Robust Temporal Monocular Depth Estimation in Unstructured Environments

Fonder

Ernst

Droogenbroeck

2022

Sensors

View full text Add to dashboard Cite

Estimating the distance to objects is crucial for autonomous vehicles, but cost, weight or power constraints sometimes prevent the use of dedicated depth sensors. In this case, the distance has to be estimated from on-board mounted RGB cameras, which is a complex task especially for environments such as natural outdoor landscapes. In this paper, we present a new depth estimation method suitable for use in such landscapes. First, we establish a bijective relationship between depth and the visual parallax of two consecutive frames and show how to exploit it to perform motion-invariant pixel-wise depth estimation. Then, we detail our architecture which is based on a pyramidal convolutional neural network where each level refines an input parallax map estimate by using two customized cost volumes. We use these cost volumes to leverage the visual spatio-temporal constraints imposed by motion and make the network robust for varied scenes. We benchmarked our approach both in test and generalization modes on public datasets featuring synthetic camera trajectories recorded in a wide variety of outdoor scenes. Results show that our network outperforms the state of the art on these datasets, while also performing well on a standard depth estimation benchmark.

show abstract

Section: Related Workmentioning

confidence: 99%

“…The approach proposed by Düzçeker et al [ 41 ] and the method called DeepV2D by Teed and Deng [ 42 ] are similar. They both propose a three-stage network.…”

Section: Related Workmentioning

confidence: 99%