DTU Benchmark Dataset
Emerging14papers using it
2021first seen
The DTU Benchmark Dataset is a collection of multi-view images used to evaluate the performance of 3D reconstruction methods, particularly in terms of accuracy and uncertainty estimation.
Papers using DTU Benchmark Dataset (14)
- Mvsnerf: Fast Generalizable Radiance Field Reconstruction From Multi-view StereoUNISURF: Unifying Neural Implicit Surfaces And Radiance Fields For Multi-view ReconstructionLearning Signed Distance Field For Multi-view Surface ReconstructionZeronvs: Zero-shot 360-degree View Synthesis From A Single ImageGc-mvsnet: Multi-view, Multi-scale, Geometrically-consistent Multi-view StereoEnhancing Multi-view Stereo With Contrastive Matching And Weighted Focal LossDepth-guided Bundle Sampling For Efficient Generalizable Neural Radiance Field ReconstructionStochastic Signed Distance ProcessesTransmvsnet: Global Context-aware Multi-view Stereo Network With TransformersRc-mvsnet: Unsupervised Multi-view Stereo With Neural RenderingMVS2D: Efficient Multi-view Stereo Via Attention-driven 2D ConvolutionsVisibility-aware Pixelwise View Selection For Multi-view Stereo MatchingMulti-View Stereo with TransformerUncertainty Quality of VGGT: An Analysis on the DTU Benchmark Dataset