Recent Neural Radiance Field (NeRF) methods on large-scale scenes have demonstrated promising results and underlined the importance of scene decomposition for scalable NeRFs. Although these methods achieved reasonable scalability, there are several critical problems remaining unexplored in the existing large-scale NeRF modeling methods, i.e., learnable decomposition, modeling scene heterogeneity, and modeling efficiency. In this paper, we introduce Switch-NeRF++, a Heterogeneous Mixture of Hash Experts (HMoHE) network that addresses these challenges within a unified framework. Our framework is a highly scalable NeRF that learns heterogeneous decomposition and heterogeneous Neural Radiance Fields efficiently for large-scale scenes in an end-to-end manner. In our framework, a gating network learns to decompose scenes into partitions and allocates 3D points to specialized NeRF experts. This gating network is co-optimized with the experts by our proposed Sparsely Gated Mixture of Experts (MoE) NeRF framework. Our network architecture incorporates a hash-based gating network and distinct heterogeneous hash experts. The hash-based gating efficiently learns the decomposition of the large-scale scene. The distinct heterogeneous hash experts consist of hash grids of different resolution ranges. This enables effective learning of the heterogeneous representation of different decomposed scene parts within large-scale complex scenes. These design choices make our framework an end-to-end and highly scalable NeRF solution for real-world large-scale scene modeling to achieve both quality and efficiency. We evaluate our accuracy and scalability on existing large-scale NeRF datasets. Additionally, we also introduce a new dataset with very large-scale scenes ($ {>} 6.5\,\text{km}^{2}$>6.5km2) from UrbanBIS. Extensive experiments demonstrate that our approach can be easily scaled to various large-scale scenes and achieve state-of-the-art scene rendering accuracy. Furthermore, our method exhibits significant efficiency gains, with an 8x acceleration in training and a 16x acceleration in rendering compared to the best-performing competitor Switch-NeRF.

This paper presents ThinkDiff, a novel alignment paradigm that empowers text-to-image diffusion models with multimodal in-context understanding and reasoning capabilities by integrating the strengths of vision-language models (VLMs). Existing multimodal diffusion finetuning methods largely focus on pixel-level reconstruction rather than in-context reasoning, and are constrained by the complexity and limited availability of reasoning-based datasets. ThinkDiff addresses these challenges by leveraging vision-language training as a proxy task, aligning VLMs with the decoder of an encoder-decoder large language model (LLM) instead of a diffusion decoder. This proxy task builds on the observation that the LLM decoder shares the same input feature space with diffusion decoders that use the corresponding LLM encoder for prompt embedding. As a result, aligning VLMs with diffusion decoders can be simplified through alignment with the LLM decoder. Without complex training and datasets, ThinkDiff effectively unleashes understanding, reasoning, and composing capabilities in diffusion models. Experiments demonstrate that ThinkDiff significantly improves accuracy from 19.2% to 46.3% on the challenging CoBSAT benchmark for multimodal in-context reasoning generation, with only 5 hours of training on 4 A100 GPUs. Additionally, ThinkDiff demonstrates exceptional performance in composing multiple images and texts into logically coherent images. Project page: https://mizhenxing.github.io/ThinkDiff.

Learning-based surface reconstruction methods have received considerable attention in recent years due to their excellent expressiveness. However, existing learning-based methods lack scalability in processing large-scale point clouds. This paper proposes a novel scalable learning-based 3D surface reconstruction method based on octree, called SSRNet. SSRNet works in a scalable reconstruction pipeline, which divides oriented point clouds into different local parts and then processes them in parallel. Accommodating this scalable design pattern, SSRNet constructs local geometric features for octree vertices. Such features comprise the relation between the vertices and the implicit surface, ensuring geometric perception. Focusing on local geometric information also enables the network to avoid the overfitting problem and generalize well on different datasets. Finally, as a learning-based method, SSRNet can process large-scale point clouds in a short time. And to further solve the efficiency problem, we provide a lightweight and efficient version that is about five times faster while maintaining reconstruction performance. Experiments show that our methods achieve state-of-the-art performance with outstanding efficiency.

Multi-view Stereo (MVS) with known camera parameters is essentially a 1D search problem within a valid depth range. Recent deep learning-based MVS methods typically densely sample depth hypotheses in the depth range, and then construct prohibitively memory-consuming 3D cost volumes for depth prediction. Although coarse-to-fine sampling strategies alleviate this overhead issue to a certain extent, the efficiency of MVS is still an open challenge. In this work, we propose a novel method for highly efficient MVS that remarkably decreases the memory footprint, meanwhile clearly advancing state-of-the-art depth prediction performance. We investigate what a search strategy can be reasonably optimal for MVS taking into account of both efficiency and effectiveness. We first formulate MVS as a binary search problem, and accordingly propose a generalized binary search network for MVS. Specifically, in each step, the depth range is split into 2 bins with extra 1 error tolerance bin on both sides. A classification is performed to identify which bin contains the true depth. We also design three mechanisms to respectively handle classification errors, deal with out-of-range samples and decrease the training memory. The new formulation makes our method only sample a very small number of depth hypotheses in each step, which is highly memory efficient, and also greatly facilitates quick training convergence. Experiments on competitive benchmarks show that our method achieves state-of-the-art accuracy with much less memory. Particularly, our method obtains an overall score of 0.289 on DTU dataset and tops the first place on challenging Tanks and Temples advanced dataset among all the learning-based methods. Our code will be released at https://github.com/MiZhenxing/GBi-Net.

In this paper, a novel learning-based network, named DeepDT, is proposed to reconstruct the surface from Delaunay triangulation of point cloud. DeepDT learns to predict inside/outside labels of Delaunay tetrahedrons directly from a point cloud and corresponding Delaunay triangulation. The local geometry features are first extracted from the input point cloud and aggregated into a graph deriving from the Delaunay triangulation. Then a graph filtering is applied on the aggregated features in order to add structural regularization to the label prediction of tetrahedrons. Due to the complicated spatial relations between tetrahedrons and the triangles, it is impossible to directly generate ground truth labels of tetrahedrons from ground truth surface. Therefore, we propose a multi-label supervision strategy which votes for the label of a tetrahedron with labels of sampling locations inside it. The proposed DeepDT can maintain abundant geometry details without generating overly complex surfaces, especially for inner surfaces of open scenes. Meanwhile, the generalization ability and time consumption of the proposed method is acceptable and competitive compared with the state-of-the-art methods. Experiments demonstrate the superior performance of the proposed DeepDT.

Existing learning-based surface reconstruction methods from point clouds are still facing challenges in terms of scalability and preservation of details on large-scale point clouds. In this paper, we propose the SSRNet, a novel scalable learning-based method for surface reconstruction. The proposed SSRNet constructs local geometry-aware features for octree vertices and designs a scalable reconstruction pipeline, which not only greatly enhances the predication accuracy of the relative position between the vertices and the implicit surface facilitating the surface reconstruction quality, but also allows dividing the point cloud and octree vertices and processing different parts in parallel for superior scalability on large-scale point clouds with millions of points. Moreover, SSRNet demonstrates outstanding generalization capability and only needs several surface data for training, much less than other learning-based reconstruction methods, which can effectively avoid overfitting. The trained model of SSRNet on one dataset can be directly used on other datasets with superior performance. Finally, the time consumption with SSRNet on a large-scale point cloud is acceptable and competitive. To our knowledge, the proposed SSRNet is the first to really bring a convincing solution to the scalability issue of the learning-based surface reconstruction methods, and is an important step to make learning-based methods competitive with respect to geometry processing methods on real-world and challenging data. Experiments show that our method achieves a breakthrough in scalability and quality compared with state-of-the-art learning-based methods.