CURVATURE-GUIDED DYNAMIC SCALE NETWORKS FOR MULTI-VIEW STEREO

Authors: Khang Truong Giang, Soohwan Song, Sungho Jo

ICLR 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments showed that the proposed method outperforms other methods on complex outdoor scenes. It significantly improves the completeness of reconstructed models.
Researcher Affiliation Academia a School of Computing, KAIST, Daejeon, 34141, Republic of Korea b Intelligent Robotics Research Division, ETRI, Daejeon 34129, Republic of Korea
Pseudocode No The paper describes algorithms and methods in text and uses figures to illustrate architectures, but it does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes Our source code is available at https://github.com/ Truong Khang/cds-mvsnet.
Open Datasets Yes The datasets used in our evaluation are DTU (Aanæs et al., 2016), Blended MVS (Yao et al., 2020), and Tanks & Temples (Knapitsch et al., 2017).
Dataset Splits Yes The evaluation of depth map and pointcloud is collected on DTU validation and test set, respectively.
Hardware Specification Yes The entire network CDS-MVSNet was trained with 30 epochs, using the SGD optimizer with an initial learning rate of 0.01, and on two NVIDIA Titan V GPU with a batch size of 6.
Software Dependencies No The paper mentions using the 'SGD optimizer' but does not specify versions for any software libraries or dependencies (e.g., Python, PyTorch, TensorFlow, CUDA).
Experiment Setup Yes For CDS-MVSNet, we set the number of input views to 3, the number of depth hypothesis planes to {48, 32, 8} with the corresponding depth interval scales {4, 2, 1}. To train CDFSNet effectively, we used the feature loss as mentioned in Section 4.2. Besides, we implemented the scale selection step of CDSConv by first initializing the Softmax temperature by 1. This hyperparameter was then decreased over training time until it reached 0.01 to produce the sparse output for scale selection. The entire network CDS-MVSNet was trained with 30 epochs, using the SGD optimizer with an initial learning rate of 0.01, and on two NVIDIA Titan V GPU with a batch size of 6.