Matching cost aggregation plays a fundamental role in learning-based multi-view stereo networks. However, directly aggregating adjacent costs can lead to suboptimal results due to local geometric inconsistency. Related methods either seek selective aggregation or improve aggregated depth in the 2D space, both are unable to handle geometric inconsistency in the cost volume effectively. In this paper, we propose GoMVS to aggregate geometrically consistent costs, yielding better utilization of adjacent geometries. More specifically, we correspond and propagate adjacent costs to the reference pixel by leveraging the local geometric smoothness in conjunction with surface normals. We achieve this by the geometric consistent propagation (GCP) module. It computes the correspondence from the adjacent depth hypothesis space to the reference depth space using surface normals, then uses the correspondence to propagate adjacent costs to the reference geometry, followed by a convolution for aggregation. Our method achieves new state-of-the-art performance on DTU, Tanks & Temple, and ETH3D datasets. Notably, our method ranks 1st on the Tanks & Temple Advanced benchmark.
University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）；Facebook Reality Labs（Facebook现实实验室）
We propose a system that uses a convolution neural network (CNN) to estimate depth from a stereo pair followed by volumetric fusion of the predicted depth maps to produce a 3D reconstruction of a scene. Our proposed depth refinement architecture, predicts view-consistent disparity and occlusion maps that helps the fusion system to produce geometrically consistent reconstructions. We utilize 3D dilated convolutions in our proposed cost filtering network that yields better filtering while almost halving the computational cost in comparison to state of the art cost filtering architectures.For fea…
SenseTime Research（商汤科技研究院）；The Chinese University of Hong Kong（香港中文大学）；Sun Yat-sen University（中山大学）
Previous monocular depth estimation methods take a single view and directly regress the expected results. Though recent advances are made by applying geometrically inspired loss functions during training, the inference procedure does not explicitly impose any geometrical constraint. Therefore these models purely rely on the quality of data and the effectiveness of learning to generalize. This either leads to suboptimal results or the demand of huge amount of expensive ground truth labelled data to generate reasonable results. In this paper, we show for the first time that the monocular depth e…
We propose a learning-based multi-view stereo (MVS) method in scattering media, such as fog or smoke, with a novel cost volume, called the dehazing cost volume. Images captured in scattering media are degraded due to light scattering and attenuation caused by suspended particles. This degradation depends on scene depth; thus, it is difficult for traditional MVS methods to evaluate photometric consistency because the depth is unknown before three-dimensional (3D) reconstruction. The dehazing cost volume can solve this chicken-and-egg problem of depth estimation and image restoration by computin…
Rethinking Monocular Depth Embedding for Generalized Stereo Matching
Shaanxi Key Laboratory for Network Computing and Security Technology, Xi’an University of Technology（陕西理工大学网络计算与安全技术重点实验室）；College of Computer Science, Sichuan University（四川大学计算机学院）
360-GeoGS: Geometrically Consistent Feed-Forward 3D Gaussian Splatting Reconstruction for 360 Images
Shanghai Key Laboratory of Navigation and Location Based Services（上海导航与基于位置的服务关键实验室）；Shanghai Jiao Tong University（上海交通大学）；State Key Laboratory of Submarine Geoscience（海底地球科学国家重点实验室）；School of Automation and Intelligent Sensing（自动化与智能感知学院）
ETH Zurich（苏黎世联邦理工学院）；Microsoft Mixed Reality & AI Zurich Lab（微软苏黎世混合现实与人工智能实验室）
We propose an online multi-view depth prediction approach on posed video streams, where the scene geometry information computed in the previous time steps is propagated to the current time step in an efficient and geometrically plausible way. The backbone of our approach is a real-time capable, lightweight encoder-decoder that relies on cost volumes computed from pairs of images. We extend it by placing a ConvLSTM cell at the bottleneck layer, which compresses an arbitrary amount of past information in its states. The novelty lies in propagating the hidden state of the cell by accounting for t…
Progressive Per-Branch Depth Optimization for DEFOM-Stereo and SAM3 Joint Analysis in UAV Forestry Applications
Centre for Data Science and Artificial Intelligence（数据科学与人工智能中心）；Victoria University of Wellington（威灵顿维多利亚大学）；Department of Computer Science and Software Engineering（计算机科学与软件工程系）；University of Canterbury（坎特伯雷大学）
Single-view depth estimation (SVDE) plays a crucial role in scene understanding for AR applications, 3D modeling, and robotics, providing the geometry of a scene based on a single image. Recent works have shown that a successful solution strongly relies on the diversity and volume of training data. This data can be sourced from stereo movies and photos. However, they do not provide geometrically complete depth maps (as disparities contain unknown shift value). Therefore, existing models trained on this data are not able to recover correct 3D representations. Our work shows that a model trained…
Blending 3D Geometry and Machine Learning for Multi-View Stereopsis
Luddy School of Informatics, Computing, and Engineering, Indiana University（信息学、计算与工程学院，印第安纳大学）；Department of Mathematics and Computer Science, Drake University（数学与计算机科学系，德雷克大学）；Electronics and Telecommunications Research Institute（电子与电信研究所）
提出GC MVSNet++，在学习阶段主动强制多视图多尺度几何一致性，并设计密集连接的代价正则化网络，在DTU和BlendedMVS上达最优，Tanks and Temples上获第二。
University of Maryland（马里兰大学）；Microsoft Research（微软研究院）
Recent work has shown that the structure of convolutional neural networks (CNNs) induces a strong prior that favors natural images. This prior, known as a deep image prior (DIP), is an effective regularizer in inverse problems such as image denoising and inpainting. We extend the concept of the DIP to depth images. Given color images and noisy and incomplete target depth maps, we optimize a randomly-initialized CNN model to reconstruct a depth map restored by virtue of using the CNN network structure as a prior combined with a view-constrained photo-consistency loss. This loss is computed usin…
A method for learning local affine-covariant regions is presented. We show that maximizing geometric repeatability does not lead to local regions, a.k.a features,that are reliably matched and this necessitates descriptor-based learning. We explore factors that influence such learning and registration: the loss function, descriptor type, geometric parametrization and the trade-off between matchability and geometric accuracy and propose a novel hard negative-constant loss function for learning of affine regions. The affine shape estimator -- AffNet -- trained with the hard negative-constant loss…
OracleGS: Grounding Generative Priors for Sparse-View Gaussian Splatting
ETH Zürich（苏黎世联邦理工学院）；Koç University（科卡大学）；KUIS AI Center（KUIS人工智能中心）；Technical University of Munich（慕尼黑技术大学）；Google（谷歌）；Munich Center for Machine Learning（慕尼黑机器学习中心）
