Bidirectional Hybrid LSTM Based Recurrent Neural Network for Multi-View Stereo

Recently, deep learning based multi-view stereo (MVS) networks have demonstrated their excellent performance on various benchmarks. In this paper, we present an effective and efficient recurrent neural network (RNN) for accurate and complete dense point cloud reconstruction. Instead of regularizing...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE transactions on visualization and computer graphics Ročník 30; číslo 7; s. 3062 - 3073
Hlavní autori: Wei, Zizhuang, Zhu, Qingtian, Min, Chen, Chen, Yisong, Wang, Guoping
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States IEEE 01.07.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:
3D reconstruction
Costs
deep learning
Feature extraction
Image reconstruction
Machine learning
multi-view stereo
Point cloud compression
point clouds
recurrent neural network
Recurrent neural networks
Regularization
Run time (computers)
Runtime
Three dimensional models
Three-dimensional displays
ISSN:1077-2626, 1941-0506, 1941-0506
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Recently, deep learning based multi-view stereo (MVS) networks have demonstrated their excellent performance on various benchmarks. In this paper, we present an effective and efficient recurrent neural network (RNN) for accurate and complete dense point cloud reconstruction. Instead of regularizing the cost volume via conventional 3D CNN or unidirectional RNN like previous attempts, we adopt a bidirectional hybrid Long Short-Term Memory (LSTM) based structure for cost volume regularization. The proposed bidirectional recurrent regularization is able to perceive full-space context information comparable to 3D CNNs while saving runtime memory. For post-processing, we introduce a visibility based approach for depth map refinement to obtain more accurate dense point clouds. Extensive experiments on DTU, Tanks and Temples and ETH3D datasets demonstrate that our method outperforms previous state-of-the-art MVS methods and exhibits high memory efficiency at runtime.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1077-2626
1941-0506
1941-0506
DOI:10.1109/TVCG.2022.3165860